1 # NOTE: this is from original KAME distribution.
2 # Some portion of this document is not applicable to the code merged into
3 # FreeBSD-current (for example, section 5).
9 $KAME: IMPLEMENTATION,v 1.216 2001/05/25 07:43:01 jinmei Exp $
10 $FreeBSD: src/share/doc/IPv6/IMPLEMENTATION,v 1.1.2.2 2001/07/03 11:01:21 ume Exp $
16 The KAME kit conforms, or tries to conform, to the latest set of IPv6
17 specifications. For future reference we list some of the relevant documents
18 below (NOTE: this is not a complete list - this is too hard to maintain...).
19 For details please refer to specific chapter in the document, RFCs, manpages
20 come with KAME, or comments in the source code.
22 Conformance tests have been performed on past and latest KAME STABLE kit,
23 at TAHI project. Results can be viewed at http://www.tahi.org/report/KAME/.
24 We also attended Univ. of New Hampshire IOL tests (http://www.iol.unh.edu/)
25 in the past, with our past snapshots.
27 RFC1639: FTP Operation Over Big Address Records (FOOBAR)
28 * RFC2428 is preferred over RFC1639. ftp clients will first try RFC2428,
29 then RFC1639 if failed.
30 RFC1886: DNS Extensions to support IPv6
31 RFC1933: (see RFC2893)
32 RFC1981: Path MTU Discovery for IPv6
33 RFC2080: RIPng for IPv6
34 * KAME-supplied route6d, bgpd and hroute6d support this.
35 RFC2283: Multiprotocol Extensions for BGP-4
37 * KAME-supplied bgpd supports this.
38 RFC2292: Advanced Sockets API for IPv6
39 * For supported library functions/kernel APIs, see sys/netinet6/ADVAPI.
40 RFC2362: Protocol Independent Multicast-Sparse Mode (PIM-SM)
41 * RFC2362 defines the packet formats and the protcol of PIM-SM.
42 RFC2373: IPv6 Addressing Architecture
43 * KAME supports node required addresses, and conforms to the scope
45 RFC2374: An IPv6 Aggregatable Global Unicast Address Format
46 * KAME supports 64-bit length of Interface ID.
47 RFC2375: IPv6 Multicast Address Assignments
48 * Userland applications use the well-known addresses assigned in the RFC.
49 RFC2428: FTP Extensions for IPv6 and NATs
50 * RFC2428 is preferred over RFC1639. ftp clients will first try RFC2428,
51 then RFC1639 if failed.
52 RFC2460: IPv6 specification
53 RFC2461: Neighbor discovery for IPv6
54 * See 1.2 in this document for details.
55 RFC2462: IPv6 Stateless Address Autoconfiguration
56 * See 1.4 in this document for details.
57 RFC2463: ICMPv6 for IPv6 specification
58 * See 1.8 in this document for details.
59 RFC2464: Transmission of IPv6 Packets over Ethernet Networks
60 RFC2465: MIB for IPv6: Textual Conventions and General Group
61 * Necessary statistics are gathered by the kernel. Actual IPv6 MIB
62 support is provided as patchkit for ucd-snmp.
63 RFC2466: MIB for IPv6: ICMPv6 group
64 * Necessary statistics are gathered by the kernel. Actual IPv6 MIB
65 support is provided as patchkit for ucd-snmp.
66 RFC2467: Transmission of IPv6 Packets over FDDI Networks
67 RFC2472: IPv6 over PPP
68 RFC2492: IPv6 over ATM Networks
69 * only PVC is supported.
70 RFC2497: Transmission of IPv6 packet over ARCnet Networks
71 RFC2545: Use of BGP-4 Multiprotocol Extensions for IPv6 Inter-Domain Routing
72 RFC2553: Basic Socket Interface Extensions for IPv6
73 * IPv4 mapped address (3.7) and special behavior of IPv6 wildcard bind
75 - supported and turned on by default on KAME/FreeBSD[34]x
77 - supported but turned off by default on KAME/NetBSD,
78 - not supported on KAME/FreeBSD228, KAME/OpenBSD and KAME/BSDI3.
79 see 1.12 in this document for details.
80 RFC2671: Extension Mechanisms for DNS (EDNS0)
81 * see USAGE for how to use it.
82 * not supported on kame/freebsd4 and kame/bsdi4.
83 RFC2673: Binary Labels in the Domain Name System
84 * KAME/bsdi4 supports A6, DNAME and binary label to some extent.
85 * KAME apps/bind8 repository has resolver library with partial A6, DNAME
86 and binary label support.
87 RFC2675: IPv6 Jumbograms
88 * See 1.7 in this document for details.
89 RFC2710: Multicast Listener Discovery for IPv6
90 RFC2711: IPv6 router alert option
91 RFC2732: Format for Literal IPv6 Addresses in URL's
92 * The spec is implemented in programs that handle URLs
93 (like freebsd ftpio(3) and fetch(1), or netbsd ftp(1))
94 RFC2766: Network Address Translation - Protocol Translation (NAT-PT)
95 * Section 4.2 is implemented by totd (see ports/totd, or pkgsrc/net/totd).
96 RFC2874: DNS Extensions to Support IPv6 Address Aggregation and Renumbering
97 * KAME/bsdi4 supports A6, DNAME and binary label to some extent.
98 * KAME apps/bind8 repository has resolver library with partial A6, DNAME
99 and binary label support.
100 RFC2893: Transition Mechanisms for IPv6 Hosts and Routers
101 * IPv4 compatible address is not supported.
102 * automatic tunneling (4.3) is not supported.
103 * "gif" interface implements IPv[46]-over-IPv[46] tunnel in a generic way,
104 and it covers "configured tunnel" described in the spec.
105 See 1.5 in this document for details.
106 RFC2894: Router renumbering for IPv6
107 RFC3041: Privacy Extensions for Stateless Address Autoconfiguration in IPv6
108 RFC3056: Connection of IPv6 Domains via IPv4 Clouds
110 * "stf" interface implements it. Be sure to read
111 draft-itojun-ipv6-transition-abuse-01.txt
112 below before configuring it, there can be security issues.
113 draft-ietf-ipngwg-icmp-name-lookups-07: IPv6 Name Lookups Through ICMP
114 draft-ietf-dhc-dhcpv6-15.txt: DHCPv6
115 draft-ietf-dhc-dhcpv6exts-12.txt: Extensions for DHCPv6
116 * kame/dhcp6 has test implementation, which will not be compiled in
118 * 15/12 drafts are not explicit about padding and string termination.
119 at IETF48, the author confirmed that there's no padding/termination
120 (and extensions can appear unaligned). our code follows the comment.
121 draft-itojun-ipv6-tcp-to-anycast-00.txt:
122 Disconnecting TCP connection toward IPv6 anycast address
123 draft-ietf-ipngwg-rfc2553bis-03.txt:
124 Basic Socket Interface Extensions for IPv6 (revised)
125 draft-ietf-ipngwg-rfc2292bis-02.txt:
126 Advanced Sockets API for IPv6 (revised)
127 * Some of the updates in the draft are not implemented yet. See
128 TODO.2292bis for more details.
129 draft-ietf-mobileip-ipv6-13.txt: Mobility Support in IPv6
131 draft-ietf-ngtrans-tcpudp-relay-04.txt:
132 An IPv6-to-IPv4 transport relay translator
133 * FAITH tcp relay translator (faithd) implements this. See 3.1 for more
135 draft-ietf-ipngwg-router-selection-01.txt:
136 Default Router Preferences and More-Specific Routes
138 draft-ietf-ipngwg-scoping-arch-02.txt:
139 The architecture, text representation, and usage of IPv6
141 * some part of the documentation (especially about the routing
142 model) is not supported yet.
143 draft-ietf-pim-sm-v2-new-02.txt
144 A revised version of RFC2362, which includes the IPv6 specific
145 packet format and protocol descriptions.
146 draft-ietf-dnsext-mdns-00.txt: Multicast DNS
147 * kame/mdnsd has test implementation, which will not be built in
148 default compilation. The draft will experience a major change in the
149 near future, so don't rely upon it.
150 draft-itojun-ipv6-transition-abuse-02.txt:
151 Possible abuse against IPv6 transition technologies (expired)
152 * KAME does not implement RFC1933/2893 automatic tunnel.
153 * "stf" interface implements some address filters. Refer to stf(4)
154 for details. Since there's no way to make 6to4 interface 100% secure,
155 we do not include "stf" interface into GENERIC.v6 compilation.
156 * kame/openbsd completely disables IPv4 mapped address support.
157 * kame/netbsd makes IPv4 mapped address support off by default.
158 * See section 1.12.6 and 1.14 for more details.
159 draft-itojun-ipv6-tclass-api-02.txt: Socket API for IPv6 traffic class field
160 draft-itojun-ipv6-flowlabel-api-01.txt: Socket API for IPv6 flow label field
161 * no consideration is made against the use of routing headers and such.
163 1.2 Neighbor Discovery
165 Neighbor Discovery is fairly stable. Currently Address Resolution,
166 Duplicated Address Detection, and Neighbor Unreachability Detection
167 are supported. In the near future we will be adding Unsolicited Neighbor
168 Advertisement transmission command as admin tool.
170 Duplicated Address Detection (DAD) will be performed when an IPv6 address
171 is assigned to a network interface, or the network interface is enabled
172 (ifconfig up). It is documented in RFC2462 5.4.
173 If DAD fails, the address will be marked "duplicated" and message will be
174 generated to syslog (and usually to console). The "duplicated" mark
175 can be checked with ifconfig. It is administrators' responsibility to check
176 for and recover from DAD failures. We may try to improve failure recovery
178 DAD procedure may not be effective on certain network interfaces/drivers.
179 If a network driver needs long initialization time (with wireless network
180 interfaces this situation is popular), and the driver mistakingly raises
181 IFF_RUNNING before the driver becomes ready, DAD code will try to transmit
182 DAD probes to not-really-ready network driver and the packet will not go out
183 from the interface. In such cases, network drivers should be corrected.
185 Some of network drivers loop multicast packets back to themselves,
186 even if instructed not to do so (especially in promiscuous mode).
187 In such cases DAD may fail, because DAD engine sees inbound NS packet
188 (actually from the node itself) and considers it as a sign of duplicate.
189 In this case, drivers should be corrected to honor IFF_SIMPLEX behavior.
190 For example, you may need to check source MAC address on a inbound packet,
191 and reject it if it is from the node itself.
192 You may also want to look at #if condition marked "heuristics" in
193 sys/netinet6/nd6_nbr.c:nd6_dad_timer() as workaround (note that the code
194 fragment in "heuristics" section is not spec conformant).
196 Neighbor Discovery specification (RFC2461) does not talk about neighbor
197 cache handling in the following cases:
198 (1) when there was no neighbor cache entry, node received unsolicited
199 RS/NS/NA/redirect packet without link-layer address
200 (2) neighbor cache handling on medium without link-layer address
201 (we need a neighbor cache entry for IsRouter bit)
202 For (1), we implemented workaround based on discussions on IETF ipngwg mailing
203 list. For more details, see the comments in the source code and email
204 thread started from (IPng 7155), dated Feb 6 1999.
206 IPv6 on-link determination rule (RFC2461) is quite different from assumptions
207 in BSD IPv4 network code. To implement behavior in RFC2461 section 5.2
208 (when default router list is empty), the kernel needs to know the default
209 outgoing interface. To configure the default outgoing interface, use
210 commands like "ndp -I de0" as root. Note that the spec misuse the word
211 "host" and "node" in several places in the section.
213 To avoid possible DoS attacks and infinite loops, KAME stack will accept
214 only 10 options on ND packet. Therefore, if you have 20 prefix options
215 attached to RA, only the first 10 prefixes will be recognized.
216 If this troubles you, please contact KAME team and/or modify
217 nd6_maxndopt in sys/netinet6/nd6.c. If there are high demands we may
218 provide sysctl knob for the variable.
220 Proxy Neighbor Advertisement support is implemented in the kernel.
221 For instance, you can configure it by using the following command:
222 # ndp -s fe80::1234%ne0 0:1:2:3:4:5 proxy
223 where ne0 is the interface which attaches to the same link as the
225 There are certain limitations, though:
226 - It does not send unsolicited multicast NA on configuration. This is MAY
228 - It does not add random delay before transmission of solicited NA. This is
229 SHOULD behavior in RFC2461.
230 - We cannot configure proxy NDP for off-link address. The target address for
231 proxying must be link-local address, or must be in prefixes configured to
232 node which does proxy NDP.
233 - RFC2461 is unclear about if it is legal for a host to perform proxy ND.
234 We do not prohibit hosts from doing proxy ND, but there will be very limited
237 Starting mid March 2000, we support Neighbor Unreachability Detection (NUD)
238 on p2p interfaces, including tunnel interfaces (gif). NUD is turned on by
239 default. Before March 2000 KAME stack did not perform NUD on p2p interfaces.
240 If the change raises any interoperability issues, you can turn off/on NUD
241 by per-interface basis. Use "ndp -i interface -nud" to turn it off.
242 Consult ndp(8) for details.
244 RFC2461 specifies upper-layer reachability confirmation hint. Whenever
245 upper-layer reachability confirmation hint comes, ND process can use it
246 to optimize neighbor discovery process - ND process can omit real ND exchange
247 and keep the neighbor cache state in REACHABLE.
248 We currently have two sources for hints: (1) setsockopt(IPV6_REACHCONF)
249 defined by 2292bis API, and (2) hints from tcp_input.
250 It is questionable if they are really trustworthy. For example, a rogue
251 userland program can use IPV6_REACHCONF to confuse ND process. Neighbor
252 cache is a system-wide information pool, and it is bad to allow single process
253 to affect others. Also, tcp_input can be hosed by hijack attempts. It is
254 wrong to allow hijack attempts to affect ND process.
255 Starting June 2000, ND code has a protection mechanism against incorrect
256 upper-layer reachability confirmation. ND code counts subsequent upper-layer
257 hints. If the number of hints reaches maximum, ND code will ignore further
258 upper-layer hints and run real ND process to confirm reachability to the peer.
259 sysctl net.inet6.icmp6.nd6_maxnudhint defines maximum # of subsequent
260 upper-layer hints to be accepted.
261 (from April 2000 to June 2000, we rejected setsockopt(IPV6_REACHCONF) from
262 non-root process - after local discussion, it looks that hints are not
263 that trustworthy even if they are from privileged processes)
265 If inbound ND packets carry invalid values, the KAME kernel will
266 drop these packet and increment statistics variable. See
267 "netstat -sn", icmp6 section. For detailed debugging session, you can
268 turn on syslog output from the kernel on errors, by turning on sysctl MIB
269 net.inet6.icmp6.nd6_debug. nd6_debug can be turned on at bootstrap
270 time, by defining ND6_DEBUG kernel compilation option (so you can
271 debug behavior during bootstrap). nd6_debug configuration should
272 only be used for test/debug purposes - for production environment,
273 nd6_debug must be set to 0. If you leave it to 1, malicious parties
274 can inject broken packet and fill up /var/log partition.
278 IPv6 uses scoped addresses. It is therefore very important to
279 specify the scope zone index (link index for a link-local address, or
280 site index for a site-local address) with an IPv6 address. Without a
281 zone index, a scoped IPv6 address is ambiguous to the kernel, and
282 the kernel would not be able to determine the outbound link for a
283 packet to the scoped address. KAME code tries to address the issue in
286 The entire architecture of scoped addresses is documented in
287 draft-ietf-ipngwg-scoping-arch-xx.txt. One non-trivial point of the
288 architecture is that the link scope is (theoretically) larger than the
289 interface scope. That is, two different interfaces can belong to a
290 same single link. However, in a normal operation, we can assume that
291 there is 1-to-1 relationship between links and interfaces. In
292 other words, we can usually put links and interfaces in the same scope
293 type. The current KAME implementation assumes the 1-to-1
294 relationship. In particular, we use interface names such as "ne1" as
295 unique link identifiers. This would be much more human-readable and
296 intuitive than numeric identifiers, but please keep your mind on the
297 theoretical difference between links and interfaces.
299 Site-local addresses are very vaguely defined in the specs, and both
300 the specification and the KAME code need tons of improvements to
301 enable its actual use. For example, it is still very unclear how we
302 define a site, or how we resolve host names in a site. There is work
303 underway to define behavior of routers at site border, but, we have
304 almost no code for site boundary node support (both forwarding nor
305 routing) and we bet almost noone has. We recommend, at this moment,
306 you to use global addresses for experiments - there are way too many
307 pitfalls if you use site-local addresses.
309 1.3.1 Kernel internal
311 In the kernel, the link index for a link-local scope address is
312 embedded into the 2nd 16bit-word (the 3rd and 4th bytes) in the IPv6
314 For example, you may see something like:
315 fe80:1::200:f8ff:fe01:6317
316 in the routing table and the interface address structure (struct
317 in6_ifaddr). The address above is a link-local unicast address which
318 belongs to a network link whose link identifier is 1 (note that it
319 eqauls to the interface index by the assumption of our
320 implementation). The embedded index enables us to identify IPv6
321 link-local addresses over multiple links effectively and with only a
324 1.3.2 Interaction with API
326 There are several candidates of API to deal with scoped addresses
329 The IPV6_PKTINFO ancillary data type or socket option defined in the
330 advanced API (RFC2292 or draft-ietf-ipngwg-rfc2292bis-xx) can specify
331 the outgoing interface of a packet. Similarly, the IPV6_PKTINFO or
332 IPV6_RECVPKTINFO socket options tell kernel to pass the incoming
333 interface to user applications.
335 These options are enough to disambiguate scoped addresses of an
336 incoming packet, because we can uniquely identify the corresponding
337 zone of the scoped address(es) by the incoming interface. However,
338 they are too strong for outgoing packets. For example, consider a
339 multi-sited node and suppose that more than one interface of the node
340 belongs to a same site. When we want to send a packet to the site,
341 we can only specify one of the interfaces for the outgoing packet with
342 these options; we cannot just say "send the packet to (one of the
343 interfaces of) the site."
345 Another kind of candidates is to use the sin6_scope_id member in the
346 sockaddr_in6 structure, defined in RFC2553 and
347 draft-ietf-ipngwg-rfc2553bis-xx.txt. The KAME kernel interprets the
348 sin6_scope_id field properly in order to disambiguate scoped
349 addresses. For example, if an application passes a sockaddr_in6
350 structure that has a non-zero sin6_scope_id value to the sendto(2)
351 system call, the kernel should send the packet to the appropriate zone
352 according to the sin6_scope_id field. Similarly, when the source or
353 the destination address of an incoming packet is a scoped one, the
354 kernel should detect the correct zone identifier based on the address
355 and the receiving interface, fill the identifier in the sin6_scope_id
356 field of a sockaddr_in6 structure, and then pass the packet to an
357 application via the recvfrom(2) system call, etc.
359 However, the semantics of the sin6_scope_id is still vague and on the
360 way to standardization. Additionally, not so many operating systems
361 support the behavior above at this moment.
364 - If your target system is limited to KAME based ones (i.e. BSD
365 variants and KAME snaps), use the sin6_scope_id field assuming the
366 kernel behavior described above.
367 - Otherwise, (i.e. if your program should be portable on other systems
369 + Use the advanced API to disambiguate scoped addresses of incoming
371 + To disambiguate scoped addresses of outgoing packets,
372 * if it is okay to just specify the outgoing interface, use the
373 advanced API. This would be the case, for example, when you
374 should only consider link-local addresses and your system
375 assumes 1-to-1 relationship between links and interfaces.
376 * otherwise, sorry but you lose. Please rush the IETF IPv6
377 community into standardizing the semantics of the sin6_scope_id
380 Routing daemons and configuration programs, like route6d and ifconfig,
381 will need to manipulate the "embedded" zone index. These programs use
382 routing sockets and ioctls (like SIOCGIFADDR_IN6) and the kernel API
383 will return IPv6 addresses with the 2nd 16bit-word filled in. The
384 APIs are for manipulating kernel internal structure. Programs that
385 use these APIs have to be prepared about differences in kernels
388 getaddrinfo(3) and getnameinfo(3) support an extended numeric IPv6
389 syntax, as documented in draft-ietf-ipngwg-rfc2553bis-xx.txt. You can
390 specify the outgoing link, by using the name of the outgoing interface
391 as the link, like "fe80::1%ne0" (again, note that we assume there is
392 1-to-1 relationship between links and interfaces.) This way you will
393 be able to specify a link-local scoped address without much trouble.
395 Other APIs like inet_pton(3) and inet_ntop(3) are inherently
396 unfriendly with scoped addresses, since they are unable to annotate
397 addresses with zone identifier.
399 1.3.3 Interaction with users (command line)
401 Most of user applications now support the extended numeric IPv6
402 syntax. In this case, you can specify outgoing link, by using the name
403 of the outgoing interface like "fe80::1%ne0" (sorry for the duplicated
404 notice, but please recall again that we assume 1-to-1 relationship
405 between links and interfaces). This is even the case for some
406 management tools such as route(8) or ndp(8). For example, to install
407 the IPv6 default route by hand, you can type like
408 # route add -inet6 default fe80::9876:5432:1234:abcd%ne0
409 (Although we suggest you to run dynamic routing instead of static
410 routes, in order to avoid configuration mistakes.)
412 Some applications have command line options for specifying an
413 appropriate zone of a scoped address (like "ping6 -I ne0 ff02::1" to
414 specify the outgoing interface). However, you can't always expect such
415 options. Thus, we recommend you to use the extended format described
418 In any case, when you specify a scoped address to the command line,
419 NEVER write the embedded form (such as ff02:1::1 or fe80:2::fedc),
420 which should only be used inside the kernel (see Section 1.3.1), and
421 is not supposed to work.
425 The KAME kit implements most of the IPv6 stateless address
426 autoconfiguration in the kernel.
427 Neighbor Discovery functions are implemented in the kernel as a whole.
428 Router Advertisement (RA) input for hosts is implemented in the
429 kernel. Router Solicitation (RS) output for endhosts, RS input
430 for routers, and RA output for routers are implemented in the
433 1.4.1 Assignment of link-local, and special addresses
435 IPv6 link-local address is generated from IEEE802 address (ethernet MAC
436 address). Each of interface is assigned an IPv6 link-local address
437 automatically, when the interface becomes up (IFF_UP). Also, direct route
438 for the link-local address is added to routing table.
440 Here is an output of netstat command:
443 Destination Gateway Flags Netif Expire
444 fe80::%ed0/64 link#1 UC ed0
445 fe80::%ep0/64 link#2 UC ep0
447 Interfaces that has no IEEE802 address (pseudo interfaces like tunnel
448 interfaces, or ppp interfaces) will borrow IEEE802 address from other
449 interfaces, such as ethernet interfaces, whenever possible.
450 If there is no IEEE802 hardware attached, last-resort pseudorandom value,
451 which is from MD5(hostname), will be used as source of link-local address.
452 If it is not suitable for your usage, you will need to configure the
453 link-local address manually.
455 If an interface is not capable of handling IPv6 (such as lack of multicast
456 support), link-local address will not be assigned to that interface.
457 See section 2 for details.
459 Each interface joins the solicited multicast address and the
460 link-local all-nodes multicast addresses (e.g. fe80::1:ff01:6317
461 and ff02::1, respectively, on the link the interface is attached).
462 In addition to a link-local address, the loopback address (::1) will be
463 assigned to the loopback interface. Also, ::1/128 and ff01::/32 are
464 automatically added to routing table, and loopback interface joins
465 node-local multicast group ff01::1.
467 1.4.2 Stateless address autoconfiguration on hosts
469 In IPv6 specification, nodes are separated into two categories:
470 routers and hosts. Routers forward packets addressed to others, hosts does
471 not forward the packets. net.inet6.ip6.forwarding defines whether this
472 node is a router or a host (router if it is 1, host if it is 0).
474 It is NOT recommended to change net.inet6.ip6.forwarding while the node
475 is in operation. IPv6 specification defines behavior for "host" and "router"
476 quite differently, and switching from one to another can cause serious
477 troubles. It is recommended to configure the variable at bootstrap time only.
479 The first step in stateless address configuration is Duplicated Address
480 Detection (DAD). See 1.2 for more detail on DAD.
482 When a host hears Router Advertisement from the router, a host may
483 autoconfigure itself by stateless address autoconfiguration.
484 This behavior can be controlled by net.inet6.ip6.accept_rtadv
485 (host autoconfigures itself if it is set to 1).
486 By autoconfiguration, network address prefix for the receiving interface
487 (usually global address prefix) is added. The default route is also
490 Routers periodically generate Router Advertisement packets. To
491 request an adjacent router to generate RA packet, a host can transmit
492 Router Solicitation. To generate an RS packet at any time, use the
493 "rtsol" command. The "rtsold" daemon is also available. "rtsold"
494 generates Router Solicitation whenever necessary, and it works great
495 for nomadic usage (notebooks/laptops). If one wishes to ignore Router
496 Advertisements, use sysctl to set net.inet6.ip6.accept_rtadv to 0.
498 To generate Router Advertisement from a router, use the "rtadvd" daemon.
500 Note that the IPv6 specification assumes the following items and that
501 nonconforming cases are left unspecified:
502 - Only hosts will listen to router advertisements
503 - Hosts have single network interface (except loopback)
504 This is therefore unwise to enable net.inet6.ip6.accept_rtadv on routers,
505 or multi-interface host. A misconfigured node can behave strange
506 (KAME code allows nonconforming configuration, for those who would like
507 to do some experiments).
509 To summarize the sysctl knob:
510 accept_rtadv forwarding role of the node
512 0 0 host (to be manually configured)
514 1 0 autoconfigured host
515 (spec assumes that host has single
516 interface only, autoconfigred host with
517 multiple interface is out-of-scope)
518 1 1 invalid, or experimental
519 (out-of-scope of spec)
521 See 1.2 in the document for relationship between DAD and autoconfiguration.
525 We supply a tiny DHCPv6 server/client in kame/dhcp6. However, the
526 implementation is premature (for example, this does NOT implement
527 address lease/release), and it is not in default compilation tree on
528 some platforms. If you want to do some experiment, compile it on your
531 DHCPv6 and autoconfiguration also needs more work. "Managed" and "Other"
532 bits in RA have no special effect to stateful autoconfiguration procedure
533 in DHCPv6 client program ("Managed" bit actually prevents stateless
534 autoconfiguration, but no special action will be taken for DHCPv6 client).
536 1.5 Generic tunnel interface
538 GIF (Generic InterFace) is a pseudo interface for configured tunnel.
539 Details are described in gif(4) manpage.
545 are available. Use "gifconfig" to assign physical (outer) source
546 and destination address to gif interfaces.
547 Configuration that uses same address family for inner and outer IP
548 header (v4 in v4, or v6 in v6) is dangerous. It is very easy to
549 configure interfaces and routing tables to perform infinite level
550 of tunneling. Please be warned.
552 gif can be configured to be ECN-friendly. See 4.5 for ECN-friendliness
553 of tunnels, and gif(4) manpage for how to configure.
555 If you would like to configure an IPv4-in-IPv6 tunnel with gif interface,
556 read gif(4) carefully. You may need to remove IPv6 link-local address
557 automatically assigned to the gif interface.
559 1.6 Source Address Selection
561 KAME's source address selection takes care of the following
565 - whether an address is deprecated
566 - whether an address is temporary (in terms of RFC 3041)
567 - prefix matching against the destination
569 Roughly speaking, the selection policy is as follows:
570 - always use an address that belongs to the same scope zone as the
572 - addresses that have equal or larger scope than the scope of the
573 destination are preferred.
574 - a deprecated address is not used in new communications if an
575 alternate (non-deprecated) address is available and has sufficient
577 - a temporary address (in terms of RFC 3041 privacy extension) are
578 preferred to a public address.
579 - if none of above conditions tie-breaks, addresses assigned on the
580 outgoing interface are preferred.
581 - if none of above conditions tie-breaks, one which is longest prefix
582 matching against the destination is preferred as the last resort.
584 For instance, ::1 is selected for ff01::1,
585 fe80::200:f8ff:fe01:6317%ne0 for fe80::2a0:24ff:feab:839b%ne0.
586 To see how longest-matching works, suppose that
587 3ffe:501:808:1:200:f8ff:fe01:6317 and 3ffe:2001:9:124:200:f8ff:fe01:6317
588 are given on the outgoing interface. Then the former is chosen as the
589 source for the destination 3ffe:501:800::1. Note that even if all
590 available addresses have smaller scope than the scope of the
591 destination, we choose one anyway. For example, if we have link-local
592 and site-local addresses only, we choose a site-local addresses for a
593 global destination. If the packet is going to break a site boundary,
594 the boundary router will return an ICMPv6 destination unreachable
595 error with code 2 - beyond scope of source address.
597 The precise desripction of the algorithm is quite complicated. To
598 describe the algorithm, we introduce the following notation:
600 For a given destination D,
601 samescope(D): The set of addresses that have the same scope as D.
602 largerscope(D): The set of addresses that have a larger scope than D.
603 smallerscope(D): The set of addresses that have a smaller scope than D.
605 For a given set of addresses A,
606 DEP(A): the set of deprecated addresses in A.
607 nonDEP(A): A - DEP(A).
609 For a given set of addresses A,
610 tmp(A): the set of preferred temporary-autoconfigured or
611 manually-configure addresses in A.
613 Also, the algorithm assumes that the outgoing interface for the
614 destination D is determined. We call the interface "I".
616 The algorithm is as follows. Selection proceeds step by step as
617 described; For example, if an address is selected by item 1, item 2 and
618 later are not considered at all.
620 0. If there is no address in the same scope zone as D, just give up;
621 the packet will not be sent.
622 1. If we do not prefer temporary addresses, go to 3.
623 Otherwise, and if tmp(samescope(D)) is not empty,
624 then choose an address that is on the interface I. If every
625 address is on I, or every address is on a different interface
626 from I, choose an arbitrary one provided that an address longest
627 matching against D is always preferred.
628 2. If tmp(largerscope(D)) is not empty,
629 then choose an address that has the smallest scope. If more than one
630 address has the smallest scope, choose an arbitrary one provided
631 that addresses on I are always preferred.
632 3. If nonDEP(samescope(D)) is not empty,
633 then apply the same logic as of 1.
634 4. If nonDEP(largerscope(D)) is not empty,
635 then apply the same logic as of 2.
636 5. If we do not prefer temporary addresses, go to 7.
637 Otherwise, and if tmp(DEP(samescope(D))) is not empty,
638 then choose an address that is on the interface I. If every
639 address is on I, or every address is on a different interface
640 from I, choose an arbitrary one provided that an address longest
641 matching against D is always preferred.
642 6. If tmp(DEP(largerscope(D))) is not empty,
643 then choose an address that has the smallest scope. If more than
644 one address has the smallest scope, choose an arbitrary one provided
645 that an address on I is always preferred.
646 7. If DEP(samescope(D)) is not empty,
647 then apply the same logic as of 5.
648 8. If DEP(largerscope(D)) is not empty,
649 then apply the same logic as of 6.
650 9. If we do not prefer temporary addresses, go to 11.
651 Otherwise, and if tmp(nonDEP(smallerscope(D))) is not empty,
652 then choose an address that has the largest scope. If more than
653 one address has the largest scope, choose an arbitrary one provided
654 that an address on I is always preferred.
655 10. If tmp(DEP(smallerscope(D))) is not empty,
656 then choose an address that has the largest scope. If more than
657 one address has the largest scope, choose an arbitrary one provided
658 that an address on I is always preferred.
659 11. If nonDEP(smallerscope(D)) is not empty,
660 then apply the same logic as of 9.
661 12. If DEP(smallerscope(D)) is not empty,
662 then apply the same logic as of 10.
664 There exists a document about source address selection
665 (draft-ietf-ipngwg-default-addr-select-xx.txt). KAME's algorithm
666 described above takes a similar approach to the document, but there
667 are some differences. See the document for more details.
669 There are some cases where we do not use the above rule. One
670 example is connected TCP session, and we use the address kept in TCP
671 protocol control block (tcb) as the source.
672 Another example is source address for Neighbor Advertisement.
673 Under the spec (RFC2461 7.2.2) NA's source should be the target
674 address of the corresponding NS's target. In this case we follow
675 the spec rather than the above longest-match rule.
677 If you would like to prohibit the use of deprecated address for some
678 reason, configure net.inet6.ip6.use_deprecated to 0. The issue
679 related to deprecated address is described in RFC2462 5.5.4 (NOTE:
680 there is some debate underway in IETF ipngwg on how to use
681 "deprecated" address).
685 KAME supports the Jumbo Payload hop-by-hop option used to send IPv6
686 packets with payloads longer than 65,535 octets. But since currently
687 KAME does not support any physical interface whose MTU is more than
688 65,535, such payloads can be seen only on the loopback interface(i.e.
691 If you want to try jumbo payloads, you first have to reconfigure the
692 kernel so that the MTU of the loopback interface is more than 65,535
693 bytes; add the following to the kernel configuration file:
694 options "LARGE_LOMTU" #To test jumbo payload
695 and recompile the new kernel.
697 Then you can test jumbo payloads by the ping6 command with -b and -s
698 options. The -b option must be specified to enlarge the size of the
699 socket buffer and the -s option specifies the length of the packet,
700 which should be more than 65,535. For example, type as follows;
701 % ping6 -b 70000 -s 68000 ::1
703 The IPv6 specification requires that the Jumbo Payload option must not
704 be used in a packet that carries a fragment header. If this condition
705 is broken, an ICMPv6 Parameter Problem message must be sent to the
706 sender. KAME kernel follows the specification, but you cannot usually
707 see an ICMPv6 error caused by this requirement.
709 If KAME kernel receives an IPv6 packet, it checks the frame length of
710 the packet and compares it to the length specified in the payload
711 length field of the IPv6 header or in the value of the Jumbo Payload
712 option, if any. If the former is shorter than the latter, KAME kernel
713 discards the packet and increments the statistics. You can see the
714 statistics as output of netstat command with `-s -p ip6' option:
718 1 with data size < data length
720 So, KAME kernel does not send an ICMPv6 error unless the erroneous
721 packet is an actual Jumbo Payload, that is, its packet size is more
722 than 65,535 bytes. As described above, KAME kernel currently does not
723 support physical interface with such a huge MTU, so it rarely returns an
726 TCP/UDP over jumbogram is not supported at this moment. This is because
727 we have no medium (other than loopback) to test this. Contact us if you
730 IPsec does not work on jumbograms. This is due to some specification twists
731 in supporting AH with jumbograms (AH header size influences payload length,
732 and this makes it real hard to authenticate inbound packet with jumbo payload
733 option as well as AH).
735 There are fundamental issues in *BSD support for jumbograms. We would like to
736 address those, but we need more time to finalize the task. To name a few:
737 - mbuf pkthdr.len field is typed as "int" in 4.4BSD, so it cannot hold
738 jumbogram with len > 2G on 32bit architecture CPUs. If we would like to
739 support jumbogram properly, the field must be expanded to hold 4G +
740 IPv6 header + link-layer header. Therefore, it must be expanded to at least
741 int64_t (u_int32_t is NOT enough).
742 - We mistakingly use "int" to hold packet length in many places. We need
743 to convert them into larger numeric type. It needs a great care, as we may
744 experience overflow during packet length computation.
745 - We mistakingly check for ip6_plen field of IPv6 header for packet payload
746 length in various places. We should be checking mbuf pkthdr.len instead.
747 ip6_input() will perform sanity check on jumbo payload option on input,
748 and we can safely use mbuf pkthdr.len afterwards.
749 - TCP code needs careful updates in bunch of places, of course.
751 1.8 Loop prevention in header processing
753 IPv6 specification allows arbitrary number of extension headers to
754 be placed onto packets. If we implement IPv6 packet processing
755 code in the way BSD IPv4 code is implemented, kernel stack may
756 overflow due to long function call chain. KAME sys/netinet6 code
757 is carefully designed to avoid kernel stack overflow. Because of
758 this, KAME sys/netinet6 code defines its own protocol switch
759 structure, as "struct ip6protosw" (see netinet6/ip6protosw.h).
761 In addition to this, we restrict the number of extension headers
762 (including the IPv6 header) in each incoming packet, in order to
763 prevent a DoS attack that tries to send packets with a massive number
764 of extension headers. The upper limit can be configured by the sysctl
765 value net.inet6.ip6.hdrnestlimit. In particular, if the value is 0,
766 the node will allow an arbitrary number of headers. As of writing this
767 document, the default value is 50.
769 IPv4 part (sys/netinet) remains untouched for compatibility.
770 Because of this, if you receive IPsec-over-IPv4 packet with massive
771 number of IPsec headers, kernel stack may blow up. IPsec-over-IPv6 is okay.
775 After RFC2463 was published, IETF ipngwg has decided to disallow ICMPv6 error
776 packet against ICMPv6 redirect, to prevent ICMPv6 storm on a network medium.
777 KAME already implements this into the kernel.
779 RFC2463 requires rate limitation for ICMPv6 error packets generated by a
780 node, to avoid possible DoS attacks. KAME kernel implements two rate-
781 limitation mechanisms, tunable via sysctl:
782 - Minimum time interval between ICMPv6 error packets
783 KAME kernel will generate no more than one ICMPv6 error packet,
784 during configured time interval. net.inet6.icmp6.errratelimit
785 controls the interval (default: disabled).
786 - Maximum ICMPv6 error packet-per-second
787 KAME kernel will generate no more than the configured number of
788 packets in one second. net.inet6.icmp6.errppslimit controls the
789 maximum packet-per-second value (default: 200pps)
790 Basically, we need to pick values that are suitable against the bandwidth
791 of link layer devices directly attached to the node. In some cases the
792 default values may not fit well. We are still unsure if the default value
793 is sane or not. Comments are welcome.
797 For userland programming, we support IPv6 socket API as specified in
798 RFC2553, RFC2292 and upcoming internet drafts.
800 TCP/UDP over IPv6 is available and quite stable. You can enjoy "telnet",
801 "ftp", "rlogin", "rsh", "ssh", etc. These applications are protocol
802 independent. That is, they automatically chooses IPv4 or IPv6
805 1.11 Kernel Internals
807 (*) TCP/UDP part is handled differently between operating system platforms.
808 See 1.12 for details.
810 The current KAME has escaped from the IPv4 netinet logic. While
811 ip_forward() calls ip_output(), ip6_forward() directly calls
812 if_output() since routers must not divide IPv6 packets into fragments.
814 ICMPv6 should contain the original packet as long as possible up to
815 1280. UDP6/IP6 port unreach, for instance, should contain all
816 extension headers and the *unchanged* UDP6 and IP6 headers.
817 So, all IP6 functions except TCP6 never convert network byte
818 order into host byte order, to save the original packet.
820 tcp6_input(), udp6_input() and icmp6_input() can't assume that IP6
821 header is preceding the transport headers due to extension
822 headers. So, in6_cksum() was implemented to handle packets whose IP6
823 header and transport header is not continuous. TCP/IP6 nor UDP/IP6
824 header structure don't exist for checksum calculation.
826 To process IP6 header, extension headers and transport headers easily,
827 KAME requires network drivers to store packets in one internal mbuf or
828 one or more external mbufs. A typical old driver prepares two
829 internal mbufs for 100 - 208 bytes data, however, KAME's reference
830 implementation stores it in one external mbuf.
832 "netstat -s -p ip6" tells you whether or not your driver conforms
833 KAME's requirement. In the following example, "cce0" violates the
834 requirement. (For more information, refer to Section 2.)
842 0 two or more ext mbuf
844 Each input function calls IP6_EXTHDR_CHECK in the beginning to check
845 if the region between IP6 and its header is
846 continuous. IP6_EXTHDR_CHECK calls m_pullup() only if the mbuf has
847 M_LOOP flag, that is, the packet comes from the loopback
848 interface. m_pullup() is never called for packets coming from physical
851 TCP6 reassembly makes use of IP6 header to store reassemble
852 information. IP6 is not supposed to be just before TCP6, so
853 ip6tcpreass structure has a pointer to TCP6 header. Of course, it has
854 also a pointer back to mbuf to avoid m_pullup().
856 Like TCP6, both IP and IP6 reassemble functions never call m_pullup().
858 xxx_ctlinput() calls in_mrejoin() on PRC_IFNEWADDR. We think this is
859 one of 4.4BSD implementation flaws. Since 4.4BSD keeps ia_multiaddrs
860 in in_ifaddr{}, it can't use multicast feature if the interface has no
861 unicast address. So, if an application joins to an interface and then
862 all unicast addresses are removed from the interface, the application
863 can't send/receive any multicast packets. Moreover, if a new unicast
864 address is assigned to the interface, in_mrejoin() must be called.
865 KAME's interfaces, however, have ALWAYS one link-local unicast
866 address. These extensions have thus not been implemented in KAME.
868 1.12 IPv4 mapped address and IPv6 wildcard socket
870 RFC2553 describes IPv4 mapped address (3.7) and special behavior
871 of IPv6 wildcard bind socket (3.8). The spec allows you to:
872 - Accept IPv4 connections by AF_INET6 wildcard bind socket.
873 - Transmit IPv4 packet over AF_INET6 socket by using special form of
874 the address like ::ffff:10.1.1.1.
875 but the spec itself is very complicated and does not specify how the
876 socket layer should behave.
877 Here we call the former one "listening side" and the latter one "initiating
878 side", for reference purposes.
880 Almost all KAME implementations treat tcp/udp port number space separately
881 between IPv4 and IPv6. You can perform wildcard bind on both of the address
882 families, on the same port.
884 There are some OS-platform differences in KAME code, as we use tcp/udp
885 code from different origin. The following table summarizes the behavior.
887 listening side initiating side
888 (AF_INET6 wildcard (connection to ::ffff:10.1.1.1)
889 socket gets IPv4 conn.)
891 KAME/BSDI3 not supported not supported
892 KAME/FreeBSD228 not supported not supported
893 KAME/FreeBSD3x configurable supported
895 KAME/FreeBSD4x configurable supported
897 KAME/NetBSD configurable supported
899 KAME/BSDI4 enabled supported
900 KAME/OpenBSD not supported not supported
902 The following sections will give you more details, and how you can
903 configure the behavior.
905 Comments on listening side:
907 It looks that RFC2553 talks too little on wildcard bind issue,
908 specifically on (1) port space issue, (2) failure mode, (3) relationship
909 between AF_INET/INET6 wildcard bind like ordering constraint, and (4) behavior
910 when conflicting socket is opened/closed. There can be several separate
911 interpretation for this RFC which conform to it but behaves differently.
912 So, to implement portable application you should assume nothing
913 about the behavior in the kernel. Using getaddrinfo() is the safest way.
914 Port number space and wildcard bind issues were discussed in detail
915 on ipv6imp mailing list, in mid March 1999 and it looks that there's
916 no concrete consensus (means, up to implementers). You may want to
917 check the mailing list archives.
918 We supply a tool called "bindtest" that explores the behavior of
919 kernel bind(2). The tool will not be compiled by default.
921 If a server application would like to accept IPv4 and IPv6 connections,
922 it should use AF_INET and AF_INET6 socket (you'll need two sockets).
923 Use getaddrinfo() with AI_PASSIVE into ai_flags, and socket(2) and bind(2)
924 to all the addresses returned.
925 By opening multiple sockets, you can accept connections onto the socket with
926 proper address family. IPv4 connections will be accepted by AF_INET socket,
927 and IPv6 connections will be accepted by AF_INET6 socket (NOTE: KAME/BSDI4
928 kernel sometimes violate this - we will fix it).
930 If you try to support IPv6 traffic only and would like to reject IPv4
931 traffic, always check the peer address when a connection is made toward
932 AF_INET6 listening socket. If the address is IPv4 mapped address, you may
933 want to reject the connection. You can check the condition by using
934 IN6_IS_ADDR_V4MAPPED() macro. This is one of the reasons the author of
935 the section (itojun) dislikes special behavior of AF_INET6 wildcard bind.
937 Comments on initiating side:
939 Advise to application implementers: to implement a portable IPv6 application
940 (which works on multiple IPv6 kernels), we believe that the following
941 is the key to the success:
942 - NEVER hardcode AF_INET nor AF_INET6.
943 - Use getaddrinfo() and getnameinfo() throughout the system.
944 Never use gethostby*(), getaddrby*(), inet_*() or getipnodeby*().
945 - If you would like to connect to destination, use getaddrinfo() and try
946 all the destination returned, like telnet does.
947 - Some of the IPv6 stack is shipped with buggy getaddrinfo(). Ship a minimal
948 working version with your application and use that as last resort.
950 If you would like to use AF_INET6 socket for both IPv4 and IPv6 outgoing
951 connection, you will need tweaked implementation in DNS support libraries,
952 as documented in RFC2553 6.1. KAME libinet6 includes the tweak in
953 getipnodebyname(). Note that getipnodebyname() itself is not recommended as
954 it does not handle scoped IPv6 addresses at all. For IPv6 name resolution
955 getaddrinfo() is the preferred API. getaddrinfo() does not implement the
958 When writing applications that make outgoing connections, story goes much
959 simpler if you treat AF_INET and AF_INET6 as totally separate address family.
960 {set,get}sockopt issue goes simpler, DNS issue will be made simpler. We do
961 not recommend you to rely upon IPv4 mapped address.
963 1.12.1 KAME/BSDI3 and KAME/FreeBSD228
965 The platforms do not support IPv4 mapped address at all (both listening side
966 and initiating side). AF_INET6 and AF_INET sockets are totally separated.
968 Port number space is totally separate between AF_INET and AF_INET6 sockets.
970 It should be noted that KAME/BSDI3 and KAME/FreeBSD228 are not conformant
971 to RFC2553 section 3.7 and 3.8. It is due to code sharing reasons.
973 1.12.2 KAME/FreeBSD[34]x
975 KAME/FreeBSD3x and KAME/FreeBSD4x use shared tcp4/6 code (from
976 sys/netinet/tcp*) and shared udp4/6 code (from sys/netinet/udp*).
977 They use unified inpcb/in6pcb structure.
979 1.12.2.1 KAME/FreeBSD[34]x, listening side
981 The platform can be configured to support IPv4 mapped address/special
982 AF_INET6 wildcard bind (enabled by default). There is no kernel compilation
983 option to disable it. You can enable/disable the behavior with sysctl
984 (per-node), or setsockopt (per-socket).
986 Wildcard AF_INET6 socket grabs IPv4 connection if and only if the following
987 conditions are satisfied:
988 - there's no AF_INET socket that matches the IPv4 connection
989 - the AF_INET6 socket is configured to accept IPv4 traffic, i.e.
990 getsockopt(IPV6_V6ONLY) returns 0.
994 1.12.2.2 KAME/FreeBSD[34]x, initiating side
996 KAME/FreeBSD3x supports outgoing connection to IPv4 mapped address
997 (::ffff:10.1.1.1), if the node is configured to accept IPv4 connections
1004 KAME/NetBSD uses shared tcp4/6 code (from sys/netinet/tcp*) and shared
1005 udp4/6 code (from sys/netinet/udp*). The implementation is made differently
1006 from KAME/FreeBSD[34]x. KAME/NetBSD uses separate inpcb/in6pcb structures,
1007 while KAME/FreeBSD[34]x uses merged inpcb structure.
1009 It should be noted that the default configuration of KAME/NetBSD is not
1010 conformant to RFC2553 section 3.8. It is intentionally turned off by default
1011 for security reasons.
1013 1.12.3.1 KAME/NetBSD, listening side
1015 The platform can be configured to support IPv4 mapped address/special AF_INET6
1016 wildcard bind (disabled by default). Kernel behavior can be summarized as
1018 - default: special support code will be compiled in, but is disabled by
1019 default. It can be controlled by sysctl (net.inet6.ip6.v6only),
1020 or setsockopt(IPV6_V6ONLY).
1021 - add "INET6_V6ONLY": No special support code for AF_INET6 wildcard socket
1022 will be compiled in. AF_INET6 sockets and AF_INET sockets are totally
1023 separate. The behavior is similar to what described in 1.12.1.
1025 sysctl setting will affect per-socket configuration at in6pcb creation time
1026 only. In other words, per-socket configuration will be copied from sysctl
1027 configuration at in6pcb creation time. To change per-socket behavior, you
1028 must perform setsockopt or reopen the socket. Change in sysctl configuration
1029 will not change the behavior or sockets that are already opened.
1031 Wildcard AF_INET6 socket grabs IPv4 connection if and only if the following
1032 conditions are satisfied:
1033 - there's no AF_INET socket that matches the IPv4 connection
1034 - the AF_INET6 socket is configured to accept IPv4 traffic, i.e.
1035 getsockopt(IPV6_V6ONLY) returns 0.
1037 You cannot bind(2) with IPv4 mapped address. This is a workaround for port
1038 number duplicate and other twists.
1040 1.12.3.2 KAME/NetBSD, initiating side
1042 When you initiate a connection, you can always connect to IPv4 destination
1043 over AF_INET6 socket, usin IPv4 mapped address destination (::ffff:10.1.1.1).
1044 This is enabled independently from the configuration for listening side, and
1049 KAME/BSDI4 uses NRL-based TCP/UDP stack and inpcb source code,
1050 which was derived from NRL IPv6/IPsec stack. We guess it supports IPv4 mapped
1051 address and speical AF_INET6 wildcard bind. The implementation is, again,
1052 different from other KAME/*BSDs.
1054 1.12.4.1 KAME/BSDI4, listening side
1056 NRL inpcb layer supports special behavior of AF_INET6 wildcard socket.
1057 There is no way to disable the behavior.
1059 Wildcard AF_INET6 socket grabs IPv4 connection if and only if the following
1060 condition is satisfied:
1061 - there's no AF_INET socket that matches the IPv4 connection
1063 1.12.4.2 KAME/BSDI4, initiating side
1065 KAME/BSDi4 supports connection initiation to IPv4 mapped address
1066 (like ::ffff:10.1.1.1).
1070 KAME/OpenBSD uses NRL-based TCP/UDP stack and inpcb source code,
1071 which was derived from NRL IPv6/IPsec stack.
1073 It should be noted that KAME/OpenBSD is not conformant to RFC2553 section 3.7
1074 and 3.8. It is intentionally omitted for security reasons.
1076 1.12.5.1 KAME/OpenBSD, listening side
1078 KAME/OpenBSD disables special behavior on AF_INET6 wildcard bind for
1079 security reasons (if IPv4 traffic toward AF_INET6 wildcard bind is allowed,
1080 access control will become much harder). KAME/BSDI4 uses NRL-based TCP/UDP
1081 stack as well, however, the behavior is different due to OpenBSD's security
1084 As a result the behavior of KAME/OpenBSD is similar to KAME/BSDI3 and
1085 KAME/FreeBSD228 (see 1.12.1 for more detail).
1087 1.12.5.2 KAME/OpenBSD, initiating side
1089 KAME/OpenBSD does not support connection initiation to IPv4 mapped address
1090 (like ::ffff:10.1.1.1).
1094 IPv4 mapped address support adds a big requirement to EVERY userland codebase.
1095 Every userland code should check if an AF_INET6 sockaddr contains IPv4
1096 mapped address or not. This adds many twists:
1098 - Access controls code becomes harder to write.
1099 For example, if you would like to reject packets from 10.0.0.0/8,
1100 you need to reject packets to AF_INET socket from 10.0.0.0/8,
1101 and to AF_INET6 socket from ::ffff:10.0.0.0/104.
1102 - If a protocol on top of IPv4 is defined differently with IPv6, we need to be
1103 really careful when we determine which protocol to use.
1104 For example, with FTP protocol, we can not simply use sa_family to determine
1105 FTP command sets. The following example is incorrect:
1106 if (sa_family == AF_INET)
1107 use EPSV/EPRT or PASV/PORT; /*IPv4*/
1108 else if (sa_family == AF_INET6)
1109 use EPSV/EPRT or LPSV/LPRT; /*IPv6*/
1112 The correct code, with consideration to IPv4 mapped address, would be:
1113 if (sa_family == AF_INET)
1114 use EPSV/EPRT or PASV/PORT; /*IPv4*/
1115 else if (sa_family == AF_INET6 && IPv4 mapped address)
1116 use EPSV/EPRT or PASV/PORT; /*IPv4 command set on AF_INET6*/
1117 else if (sa_family == AF_INET6 && !IPv4 mapped address)
1118 use EPSV/EPRT or LPSV/LPRT; /*IPv6*/
1121 It is too much to ask for every body to be careful like this.
1122 The problem is, we are not sure if the above code fragment is perfect for
1124 - By enabling kernel support for IPv4 mapped address (outgoing direction),
1125 servers on the kernel can be hosed by IPv6 native packet that has IPv4
1126 mapped address in IPv6 header source, and can generate unwanted IPv4 packets.
1127 draft-itojun-ipv6-transition-abuse-01.txt talks more about this scenario.
1129 Due to the above twists, some of KAME userland programs has restrictions on
1130 the use of IPv4 mapped addresses:
1131 - rshd/rlogind do not accept connections from IPv4 mapped address.
1132 This is to avoid malicious use of IPv4 mapped address in IPv6 native
1133 packet, to bypass source-address based authentication.
1134 - ftp/ftpd assume that you are on dual stack network. IPv4 mapped address
1135 will be decoded in userland, and will be passed to AF_INET sockets
1136 (in other words, ftp/ftpd do not support SIIT environment).
1138 1.12.7 Interaction with SIIT translator
1140 SIIT translator is specified in RFC2765. KAME node cannot become a SIIT
1141 translator box, nor SIIT end node (a node in SIIT cloud).
1143 To become a SIIT translator box, we need to put additional code for that.
1144 We do not have the code in our tree at this moment.
1146 There are multiple reasons that we are unable to become SIIT end node.
1147 (1) SIIT translators require end nodes in the SIIT cloud to be IPv6-only.
1148 Since we are unable to compile INET-less kernel, we are unable to become
1149 SIIT end node. (2) As presented in 1.12.6, some of our userland code assumes
1150 dual stack network. (3) KAME stack filters out IPv6 packets with IPv4
1151 mapped address in the header, to secure non-SIIT case (which is much more
1152 common). Effectively KAME node will reject any packets via SIIT translator
1153 box. See section 1.14 for more detail about the last item.
1155 There are documentation issues too - SIIT document requires very strange
1156 things. For example, SIIT document asks IPv6-only (meaning no IPv4 code)
1157 node to be able to construct IPv4 IPsec headers. If a node knows how to
1158 construct IPv4 IPsec headers, that is not an IPv6-only node, it is a dual-stack
1159 node. The requirements imposed in SIIT document contradict with the other
1160 part of the document itself.
1162 1.13 sockaddr_storage
1164 When RFC2553 was about to be finalized, there was discussion on how struct
1165 sockaddr_storage members are named. One proposal is to prepend "__" to the
1166 members (like "__ss_len") as they should not be touched. The other proposal
1167 was that don't prepend it (like "ss_len") as we need to touch those members
1168 directly. There was no clear consensus on it.
1170 As a result, RFC2553 defines struct sockaddr_storage as follows:
1171 struct sockaddr_storage {
1172 u_char __ss_len; /* address length */
1173 u_char __ss_family; /* address family */
1174 /* and bunch of padding */
1176 On the contrary, XNET draft defines as follows:
1177 struct sockaddr_storage {
1178 u_char ss_len; /* address length */
1179 u_char ss_family; /* address family */
1180 /* and bunch of padding */
1183 In December 1999, it was agreed that RFC2553bis should pick the latter (XNET)
1186 KAME kit prior to December 1999 used RFC2553 definition. KAME kit after
1187 December 1999 (including December) will conform to XNET definition,
1188 based on RFC2553bis discussion.
1190 If you look at multiple IPv6 implementations, you will be able to see
1191 both definitions. As an userland programmer, the most portable way of
1192 dealing with it is to:
1193 (1) ensure ss_family and/or ss_len are available on the platform, by using
1195 (2) have -Dss_family=__ss_family to unify all occurences (including header
1196 file) into __ss_family, or
1197 (3) never touch __ss_family. cast to sockaddr * and use sa_family like:
1198 struct sockaddr_storage ss;
1199 family = ((struct sockaddr *)&ss)->sa_family
1201 1.14 Invalid addresses on the wire
1203 Some of IPv6 transition technologies embed IPv4 address into IPv6 address.
1204 These specifications themselves are fine, however, there can be certain
1205 set of attacks enabled by these specifications. Recent speicifcation
1206 documents covers up those issues, however, there are already-published RFCs
1207 that does not have protection against those (like using source address of
1208 ::ffff:127.0.0.1 to bypass "reject packet from remote" filter).
1210 To name a few, these address ranges can be used to hose an IPv6 implementation,
1211 or bypass security controls:
1212 - IPv4 mapped address that embeds unspecified/multicast/loopback/broadcast
1213 IPv4 address (if they are in IPv6 native packet header, they are malicious)
1214 ::ffff:0.0.0.0/104 ::ffff:127.0.0.0/104
1215 ::ffff:224.0.0.0/100 ::ffff:255.0.0.0/104
1216 - 6to4 (RFC3056) prefix generated from unspecified/multicast/loopback/
1217 broadcast/private IPv4 address
1218 2002:0000::/24 2002:7f00::/24 2002:e000::/24
1219 2002:ff00::/24 2002:0a00::/24 2002:ac10::/28
1221 - IPv4 compatible address that embeds unspecified/multicast/loopback/broadcast
1222 IPv4 address (if they are in IPv6 native packet header, they are malicious).
1223 Note that, since KAME doe snot support RFC1933/2893 auto tunnels, KAME nodes
1224 are not vulnerable to these packets.
1225 ::0.0.0.0/104 ::127.0.0.0/104 ::224.0.0.0/100 ::255.0.0.0/104
1227 Also, since KAME does not support RFC1933/2893 auto tunnels, seeing IPv4
1228 compatible is very rare. You should take caution if you see those on the wire.
1230 If we see IPv6 packets with IPv4 mapped address (::ffff:0.0.0.0/96) in the
1231 header in dual-stack environment (not in SIIT environment), they indicate
1232 that someone is trying to inpersonate IPv4 peer. The packet should be dropped.
1234 IPv6 specifications do not talk very much about IPv6 unspecified address (::)
1235 in the IPv6 source address field. Clarification is in progress.
1236 Here are couple of comments:
1237 - IPv6 unspecified address can be used in IPv6 source address field, if and
1238 only if we have no legal source address for the node. The legal situations
1239 include, but may not be limited to, (1) MLD while no IPv6 address is assigned
1240 to the node and (2) DAD.
1241 - If IPv6 TCP packet has IPv6 unspecified address, it is an attack attempt.
1242 The form can be used as a trigger for TCP DoS attack. KAME code already
1244 - The following examples are seemingly illegal. It seems that there's general
1245 consensus among ipngwg for those. (1) mobile-ip6 home address option,
1246 (2) offlink packets (so routers should not forward them).
1247 KAME implmements (2) already.
1249 KAME code is carefully written to avoid such incidents. More specifically,
1250 KAME kernel will reject packets with certain source/dstination address in IPv6
1251 base header, or IPv6 routing header. Also, KAME default configuration file
1252 is written carefully, to avoid those attacks.
1254 draft-itojun-ipv6-transition-abuse-01.txt talks about more about this.
1256 1.15 Node's required addresses
1258 RFC2373 section 2.8 talks about required addresses for an IPv6
1259 node. The section talks about how KAME stack manages those required
1264 The following items are automatically assigned to the node (or the node will
1265 automatically joins the group), at bootstrap time:
1267 - All-nodes multicast addresses (ff01::1)
1269 The following items will be automatically handled when the interface becomes
1271 - Its link-local address for each interface
1272 - Solicited-node multicast address for link-local addresses
1273 - Link-local allnodes multicast address (ff02::1)
1275 The following items need to be configured manually by ifconfig(8) or prefix(8).
1276 Alternatively, these can be autoconfigured by using stateless address
1278 - Assigned unicast/anycast addresses
1279 - Solicited-Node multicast address for assigned unicast address
1281 Users can join groups by using appropriate system calls like setsockopt(2).
1285 In addition to the above, routers needs to handle the following items.
1287 The following items need to be configured manually by using ifconfig(8).
1288 o The subnet-router anycast addresses for the interfaces it is configured
1289 to act as a router on (prefix::/64)
1290 o All other anycast addresses with which the router has been configured
1292 The router will join the following multicast group when rtadvd(8) is available
1294 o All-Routers Multicast Addresses (ff02::2)
1296 Routing daemons will join appropriate multicast groups, as necessary,
1297 like ff02::9 for RIPng.
1299 Users can join groups by using appropriate system calls like setsockopt(2).
1303 Current KAME kernel implements 2292bis API, documented in
1304 draft-ietf-ipngwg-rfc2292bis-xx.txt. It also implements RFC2292 API,
1305 for backward compatibility purposes with *BSD-integrated codebase.
1306 KAME tree ships with 2292bis headers.
1307 *BSD-integrated codebase implements either RFC2292, or 2292bis, API.
1308 see "COVERAGE" document for detailed implementation status.
1310 Here are couple of issues to mention:
1311 - *BSD-integrated binaries, compiled for RFC2292, will work on KAME kernel.
1312 For example, OpenBSD 2.7 /sbin/rtsol will work on KAME/openbsd kernel.
1313 - KAME binaries, compiled using 2292bis, will not work on *BSD-integrated
1314 kenrel. For example, KAME /usr/local/v6/sbin/rtsol will not work on
1316 - 2292bis API is not compatible with RFC2292 API. 2292bis #define symbols
1317 conflict with RFC2292 symbols. Therefore, if you compile programs that
1318 assume RFC2292 API, the compilation itself goes fine, however, the compiled
1319 binary will not work correctly. The problem is not KAME issue, but API
1320 issue. For example, Solaris 8 implements 2292bis API. If you compile
1321 RFC2292-based code on Solaris 8, the binary can behave strange.
1323 There are few (or couple of) incompatible behavior in RFC2292 binary backward
1324 compatibility support in KAME tree. To enumerate:
1325 - Type 0 routing header lacks support for strict/loose bitmap.
1326 Even if we see packets with "strict" bit set, those bits will not be made
1327 visible to the userland.
1328 Background: RFC2292 document is based on RFC1883 IPv6, and it uses
1329 strict/loose bitmap. 2292bis document is based on RFC2460 IPv6, and it has
1330 no strict/loose bitmap (it was removed from RFC2460). KAME tree obeys
1331 RFC2460 IPv6, and lacks support for strict/loose bitmap.
1335 KAME requires three items to be added into the standard drivers:
1337 (1) mbuf clustering requirement. In this stable release, we changed
1338 MINCLSIZE into MHLEN+1 for all the operating systems in order to make
1339 all the drivers behave as we expect.
1341 (2) multicast. If "ifmcstat" yields no multicast group for a
1342 interface, that interface has to be patched.
1344 To avoid troubles, we suggest you to comment out the device drivers
1345 for unsupported/unnecessary cards, from the kernel configuration file.
1346 If you accidentally enable unsupported drivers, some of the userland
1347 tools may not work correctly (routing daemons are typical example).
1349 In the following sections, "official support" means that KAME developers
1350 are using that ethernet card/driver frequently.
1352 (NOTE: In the past we required all pcmcia drivers to have a call to
1353 in6_ifattach(). We have no such requirement any more)
1355 2.1 FreeBSD 2.2.x-RELEASE
1357 Here is a list of FreeBSD 2.2.x-RELEASE drivers and its conditions:
1359 driver mbuf(1) multicast(2) official support?
1379 sr looks ok ok - (**)
1381 You may want to add an invocation of "rtsol" in "/etc/pccard_ether",
1382 if you are using notebook computers and PCMCIA ethernet card.
1384 (*) These drivers are distributed with PAO (http://www.jp.freebsd.org/PAO/).
1386 (**) There was some report says that, if you make sr driver up and down and
1387 then up, the kernel may hang up. We have disabled frame-relay support from
1388 sr driver and after that this looks to be working fine. If you need
1389 frame-relay support to come back, please contact KAME developers.
1393 The following lists BSD/OS 3.x device drivers and its conditions:
1395 driver mbuf(1) multicast(2) official support?
1416 You may want to use "@insert" directive in /etc/pccard.conf to invoke
1417 "rtsol" command right after dynamic insertion of PCMCIA ethernet cards.
1421 The following table lists the network drivers we have tried so far.
1423 driver mbuf(1) multicast(2) official support?
1426 awi pcmcia/i386 ok ok -
1427 bah zbus/amiga NG(*)
1428 cnw pcmcia/i386 ok ok yes
1429 ep pcmcia/i386 ok ok -
1430 le sbus/sparc ok ok yes
1431 ne pci/i386 ok ok yes
1432 ne pcmcia/i386 ok ok yes
1433 wi pcmcia/i386 ok ok yes
1437 (*) This may need some fix, but I'm not sure what arcnet interfaces assume...
1439 2.4 FreeBSD 3.x-RELEASE
1441 Here is a list of FreeBSD 3.x-RELEASE drivers and its conditions:
1443 driver mbuf(1) multicast(2) official support?
1456 (*) These drivers are distributed with PAO as PAO3
1457 (http://www.jp.freebsd.org/PAO/).
1458 (**) there are trouble reports with multicast filter initialization.
1460 More drivers will just simply work on KAME FreeBSD 3.x-RELEASE but have not
1463 2.5 FreeBSD 4.x-RELEASE
1465 Here is a list of FreeBSD 4.x-RELEASE drivers and its conditions:
1474 Here is a list of OpenBSD 2.x drivers and its conditions:
1476 driver mbuf(1) multicast(2) official support?
1479 de pci/i386 ok ok yes
1481 le sbus/sparc ok ok yes
1482 ne pci/i386 ok ok yes
1483 ne pcmcia/i386 ok ok yes
1484 wi pcmcia/i386 ok ok yes
1486 (*) There seem to be some problem in driver, with multicast filter
1487 configuration. This happens with certain revision of chipset on the card.
1488 Should be fixed by now by workaround in sys/net/if.c, but still not sure.
1492 The following lists BSD/OS 4.x device drivers and its conditions:
1494 driver mbuf(1) multicast(2) official support?
1500 You may want to use "@insert" directive in /etc/pccard.conf to invoke
1501 "rtsol" command right after dynamic insertion of PCMCIA ethernet cards.
1503 (*) exp driver has serious conflict with KAME initialization sequence.
1504 A workaround is committed into sys/i386/pci/if_exp.c, and should be okay by now.
1508 We categorize IPv4/IPv6 translator into 4 types.
1510 Translator A --- It is used in the early stage of transition to make
1511 it possible to establish a connection from an IPv6 host in an IPv6
1512 island to an IPv4 host in the IPv4 ocean.
1514 Translator B --- It is used in the early stage of transition to make
1515 it possible to establish a connection from an IPv4 host in the IPv4
1516 ocean to an IPv6 host in an IPv6 island.
1518 Translator C --- It is used in the late stage of transition to make it
1519 possible to establish a connection from an IPv4 host in an IPv4 island
1520 to an IPv6 host in the IPv6 ocean.
1522 Translator D --- It is used in the late stage of transition to make it
1523 possible to establish a connection from an IPv6 host in the IPv6 ocean
1524 to an IPv4 host in an IPv4 island.
1526 KAME provides an TCP relay translator for category A. This is called
1527 "FAITH". We also provide IP header translator for category A.
1529 3.1 FAITH TCP relay translator
1531 FAITH system uses TCP relay daemon called "faithd" helped by the KAME kernel.
1532 FAITH will reserve an IPv6 address prefix, and relay TCP connection
1533 toward that prefix to IPv4 destination.
1535 For example, if the reserved IPv6 prefix is 3ffe:0501:0200:ffff::, and
1536 the IPv6 destination for TCP connection is 3ffe:0501:0200:ffff::163.221.202.12,
1537 the connection will be relayed toward IPv4 destination 163.221.202.12.
1539 destination IPv4 node (163.221.202.12)
1541 | IPv4 tcp toward 163.221.202.12
1542 FAITH-relay dual stack node
1544 | IPv6 TCP toward 3ffe:0501:0200:ffff::163.221.202.12
1547 faithd must be invoked on FAITH-relay dual stack node.
1549 For more details, consult kame/kame/faithd/README and
1550 draft-ietf-ngtrans-tcpudp-relay-04.txt.
1552 3.2 IPv6-to-IPv4 header translator
1558 IPsec is implemented as the following three components.
1560 (1) Policy Management
1562 (3) AH, ESP and IPComp handling in kernel
1564 Note that KAME/OpenBSD does NOT include support for KAME IPsec code,
1565 as OpenBSD team has their home-brew IPsec stack and they have no plan
1566 to replace it. IPv6 support for IPsec is, therefore, lacking on KAME/OpenBSD.
1568 http://www.netbsd.org/Documentation/network/ipsec/ has more information
1569 including usage examples.
1571 4.1 Policy Management
1573 The kernel implements experimental policy management code. There are two way
1574 to manage security policy. One is to configure per-socket policy using
1575 setsockopt(3). In this cases, policy configuration is described in
1576 ipsec_set_policy(3). The other is to configure kernel packet filter-based
1577 policy using PF_KEY interface, via setkey(8).
1579 The policy entry will be matched in order. The order of entries makes
1580 difference in behavior.
1584 The key management code implemented in this kit (sys/netkey) is a
1585 home-brew PFKEY v2 implementation. This conforms to RFC2367.
1587 The home-brew IKE daemon, "racoon" is included in the kit (kame/kame/racoon,
1588 or usr.sbin/racoon).
1589 Basically you'll need to run racoon as daemon, then setup a policy
1590 to require keys (like ping -P 'out ipsec esp/transport//use').
1591 The kernel will contact racoon daemon as necessary to exchange keys.
1593 In IKE spec, there's ambiguity about interpretation of "tunnel" proposal.
1594 For example, if we would like to propose the use of following packet:
1595 IP AH ESP IP payload
1596 some implementation proposes it as "AH transport and ESP tunnel", since
1597 this is more logical from packet construction point of view. Some
1598 implementation proposes it as "AH tunnel and ESP tunnel".
1599 Racoon follows the former route.
1600 This raises real interoperability issue. We hope this to be resolved quickly.
1602 4.3 AH and ESP handling
1604 IPsec module is implemented as "hooks" to the standard IPv4/IPv6
1605 processing. When sending a packet, ip{,6}_output() checks if ESP/AH
1606 processing is required by checking if a matching SPD (Security
1607 Policy Database) is found. If ESP/AH is needed,
1608 {esp,ah}{4,6}_output() will be called and mbuf will be updated
1609 accordingly. When a packet is received, {esp,ah}4_input() will be
1610 called based on protocol number, i.e. (*inetsw[proto])().
1611 {esp,ah}4_input() will decrypt/check authenticity of the packet,
1612 and strips off daisy-chained header and padding for ESP/AH. It is
1613 safe to strip off the ESP/AH header on packet reception, since we
1614 will never use the received packet in "as is" form.
1616 By using ESP/AH, TCP4/6 effective data segment size will be affected by
1617 extra daisy-chained headers inserted by ESP/AH. Our code takes care of
1620 Basic crypto functions can be found in directory "sys/crypto". ESP/AH
1621 transform are listed in {esp,ah}_core.c with wrapper functions. If you
1622 wish to add some algorithm, add wrapper function in {esp,ah}_core.c, and
1623 add your crypto algorithm code into sys/crypto.
1625 Tunnel mode works basically fine, but comes with the following restrictions:
1626 - You cannot run routing daemon across IPsec tunnel, since we do not model
1627 IPsec tunnel as pseudo interfaces.
1628 - Authentication model for AH tunnel must be revisited. We'll need to
1629 improve the policy management engine, eventually.
1630 - Path MTU discovery does not work across IPv6 IPsec tunnel gateway due to
1633 AH specificaton does not talk much about "multiple AH on a packet" case.
1634 We incrementally compute AH checksum, from inside to outside. Also, we
1635 treat inner AH to be immutable.
1636 For example, if we are to create the following packet:
1637 IP AH1 AH2 AH3 payload
1638 we do it incrementally. As a result, we get crypto checksums like below:
1639 AH3 has checksum against "IP AH3' payload".
1640 where AH3' = AH3 with checksum field filled with 0.
1641 AH2 has checksum against "IP AH2' AH3 payload".
1642 AH1 has checksum against "IP AH1' AH2 AH3 payload",
1643 Also note that AH3 has the smallest sequence number, and AH1 has the largest
1646 To avoid traffic analysis on shorter packets, ESP output logic supports
1647 random length padding. By setting net.inet.ipsec.esp_randpad (or
1648 net.inet6.ipsec6.esp_randpad) to positive value N, you can ask the kernel
1649 to randomly pad packets shorter than N bytes, to random length smaller than
1650 or equal to N. Note that N does not include ESP authentication data length.
1651 Also note that the random padding is not included in TCP segment
1652 size computation. Negative value will turn off the functionality.
1653 Recommeded value for N is like 128, or 256. If you use a too big number
1654 as N, you may experience inefficiency due to fragmented packtes.
1658 IPComp stands for IP payload compression protocol. This is aimed for
1659 payload compression, not the header compression like PPP VJ compression.
1660 This may be useful when you are using slow serial link (say, cell phone)
1661 with powerful CPU (well, recent notebook PCs are really powerful...).
1662 The protocol design of IPComp is very similar to IPsec, though it was
1663 defined separately from IPsec itself.
1665 Here are some points to be noted:
1666 - IPComp is treated as part of IPsec protocol suite, and SPI and
1667 CPI space is unified. Spec says that there's no relationship
1668 between two so they are assumed to be separate in specs.
1669 - IPComp association (IPCA) is kept in SAD.
1670 - It is possible to use well-known CPI (CPI=2 for DEFLATE for example),
1671 for outbound/inbound packet, but for indexing purposes one element from
1672 SPI/CPI space will be occupied anyway.
1673 - pfkey is modified to support IPComp. However, there's no official
1674 SA type number assignment yet. Portability with other IPComp
1675 stack is questionable (anyway, who else implement IPComp on UN*X?).
1676 - Spec says that IPComp output processing must be performed before AH/ESP
1677 output processing, to achieve better compression ratio and "stir" data
1678 stream before encryption. The most meaningful processing order is:
1679 (1) compress payload by IPComp, (2) encrypt payload by ESP, then (3) attach
1680 authentication data by AH.
1681 However, with manual SPD setting, you are able to violate the ordering
1682 (KAME code is too generic, maybe). Also, it is just okay to use IPComp
1683 alone, without AH/ESP.
1684 - Though the packet size can be significantly decreased by using IPComp, no
1685 special consideration is made about path MTU (spec talks nothing about MTU
1686 consideration). IPComp is designed for serial links, not ethernet-like
1688 - You can change compression ratio on outbound packet, by changing
1689 deflate_policy in sys/netinet6/ipcomp_core.c. You can also change outbound
1690 history buffer size by changing deflate_window_out in the same source code.
1691 (should it be sysctl accessible, or per-SAD configurable?)
1692 - Tunnel mode IPComp is not working right. KAME box can generate tunnelled
1693 IPComp packet, however, cannot accept tunneled IPComp packet.
1694 - You can negotiate IPComp association with racoon IKE daemon.
1695 - KAME code does not attach Adler32 checksum to compressed data.
1696 see ipsec wg mailing list discussion in Jan 2000 for details.
1698 4.5 Conformance to RFCs and IDs
1700 The IPsec code in the kernel conforms (or, tries to conform) to the
1701 following standards:
1702 "old IPsec" specification documented in rfc182[5-9].txt
1703 "new IPsec" specification documented in:
1704 rfc240[1-6].txt rfc241[01].txt rfc2451.txt
1705 draft-mcdonald-simple-ipsec-api-01.txt
1706 (expired, available in ftp://ftp.kame.net/pub/internet-drafts/)
1707 draft-ietf-ipsec-ciph-aes-cbc-00.txt
1709 RFC2393: IP Payload Compression Protocol (IPComp)
1710 IKE specifications (rfc240[7-9].txt) are implemented in userland
1711 as "racoon" IKE daemon.
1713 Currently supported algorithms are:
1715 null crypto checksum (no document, just for debugging)
1716 keyed MD5 with 128bit crypto checksum (rfc1828.txt)
1717 keyed SHA1 with 128bit crypto checksum (no document)
1718 HMAC MD5 with 128bit crypto checksum (rfc2085.txt)
1719 HMAC SHA1 with 128bit crypto checksum (no document)
1721 null encryption (no document, similar to rfc2410.txt)
1722 DES-CBC mode (rfc1829.txt)
1724 null crypto checksum (no document, just for debugging)
1725 keyed MD5 with 96bit crypto checksum (no document)
1726 keyed SHA1 with 96bit crypto checksum (no document)
1727 HMAC MD5 with 96bit crypto checksum (rfc2403.txt
1728 HMAC SHA1 with 96bit crypto checksum (rfc2404.txt)
1729 HMAC SHA2-256 with 96bit crypto checksum (no document)
1730 HMAC SHA2-384 with 96bit crypto checksum (no document)
1731 HMAC SHA2-512 with 96bit crypto checksum (no document)
1733 null encryption (rfc2410.txt)
1734 DES-CBC with derived IV
1735 (draft-ietf-ipsec-ciph-des-derived-01.txt, draft expired)
1736 DES-CBC with explicit IV (rfc2405.txt)
1737 3DES-CBC with explicit IV (rfc2451.txt)
1738 BLOWFISH CBC (rfc2451.txt)
1739 CAST128 CBC (rfc2451.txt)
1740 RIJNDAEL/AES CBC (draft-ietf-ipsec-ciph-aes-cbc-00.txt,
1741 uses IANA-assigned protocol number)
1742 TWOFISH CBC (draft-ietf-ipsec-ciph-aes-cbc-00.txt)
1743 each of the above can be combined with:
1744 ESP authentication with HMAC-MD5(96bit)
1745 ESP authentication with HMAC-SHA1(96bit)
1747 RFC2394: IP Payload Compression Using DEFLATE
1749 The following algorithms are NOT supported:
1751 HMAC MD5 with 128bit crypto checksum + 64bit replay prevention
1753 keyed SHA1 with 160bit crypto checksum + 32bit padding (rfc1852.txt)
1755 The key/policy management API is based on the following document, with fair
1756 amount of extensions:
1757 RFC2367: PF_KEY key management API
1759 4.6 ECN consideration on IPsec tunnels
1761 KAME IPsec implements ECN-friendly IPsec tunnel, described in
1762 draft-ietf-ipsec-ecn-02.txt.
1763 Normal IPsec tunnel is described in RFC2401. On encapsulation,
1764 IPv4 TOS field (or, IPv6 traffic class field) will be copied from inner
1765 IP header to outer IP header. On decapsulation outer IP header
1766 will be simply dropped. The decapsulation rule is not compatible
1767 with ECN, since ECN bit on the outer IP TOS/traffic class field will be
1769 To make IPsec tunnel ECN-friendly, we should modify encapsulation
1770 and decapsulation procedure. This is described in
1771 draft-ietf-ipsec-ecn-02.txt, chapter 3.3.
1773 KAME IPsec tunnel implementation can give you three behaviors, by setting
1774 net.inet.ipsec.ecn (or net.inet6.ipsec6.ecn) to some value:
1775 - RFC2401: no consideration for ECN (sysctl value -1)
1776 - ECN forbidden (sysctl value 0)
1777 - ECN allowed (sysctl value 1)
1778 Note that the behavior is configurable in per-node manner, not per-SA manner
1779 (draft-ietf-ipsec-ecn-02 wants per-SA configuration, but it looks too much
1782 The behavior is summarized as follows (see source code for more detail):
1784 encapsulate decapsulate
1786 RFC2401 copy all TOS bits drop TOS bits on outer
1787 from inner to outer. (use inner TOS bits as is)
1789 ECN forbidden copy TOS bits except for ECN drop TOS bits on outer
1790 (masked with 0xfc) from inner (use inner TOS bits as is)
1791 to outer. set ECN bits to 0.
1793 ECN allowed copy TOS bits except for ECN use inner TOS bits with some
1794 CE (masked with 0xfe) from change. if outer ECN CE bit
1795 inner to outer. is 1, enable ECN CE bit on
1796 set ECN CE bit to 0. the inner.
1798 General strategy for configuration is as follows:
1799 - if both IPsec tunnel endpoint are capable of ECN-friendly behavior,
1800 you'd better configure both end to "ECN allowed" (sysctl value 1).
1801 - if the other end is very strict about TOS bit, use "RFC2401"
1803 - in other cases, use "ECN forbidden" (sysctl value 0).
1804 The default behavior is "ECN forbidden" (sysctl value 0).
1806 For more information, please refer to:
1807 draft-ietf-ipsec-ecn-02.txt
1808 RFC2481 (Explicit Congestion Notification)
1809 KAME sys/netinet6/{ah,esp}_input.c
1811 (Thanks goes to Kenjiro Cho <kjc@csl.sony.co.jp> for detailed analysis)
1813 4.7 Interoperability
1815 IPsec, IPComp (in kernel) and IKE (in userland as "racoon") has been tested
1816 at several interoperability test events, and it is known to interoperate
1817 with many other implementations well. Also, KAME IPsec has quite wide
1818 coverage for IPsec crypto algorithms documented in RFC (we do not cover
1819 algorithms with intellectual property issues, though).
1821 Here are (some of) platforms we have tested IPsec/IKE interoperability
1822 in the past, no particular order. Note that both ends (KAME and
1823 others) may have modified their implementation, so use the following
1824 list just for reference purposes.
1825 ACC, allied-telesis, Altiga, Ashley-laurent (vpcom.com), BlueSteel,
1826 CISCO IOS, Cryptek, Checkpoint FW-1, Data Fellows (F-Secure),
1827 Ericsson, Fitel, FreeS/WAN, HiFn, HITACHI, IBM AIX, IIJ, Intel Canada,
1828 Intel Packet Protect, MEW NetCocoon, MGCS, Microsoft WinNT/2000,
1829 NAI PGPnet, NetLock, NIST (linux IPsec + plutoplus), NEC IX5000,
1830 Netscreen, NxNetworks, OpenBSD isakmpd, Pivotal, Radguard, RapidStream,
1831 RedCreek, Routerware, RSA, SSH (both IPv4/IPv6), Secure Computing,
1832 Soliton, Sun Solaris8, TIS/NAI Gauntret, Toshiba, VPNet,
1835 Here are (some of) platforms we have tested IPComp/IKE interoperability
1836 in the past, in no particular order.
1837 IRE, SSH (both IPv4/IPv6), NetLock
1839 VPNC (vpnc.org) provides IPsec conformance tests, using KAME and OpenBSD
1840 IPsec/IKE implementations. Their test results are available at
1841 http://www.vpnc.org/conformance.html, and it may give you more idea
1842 about which implementation interoperates with KAME IPsec/IKE implementation.
1846 KAME kit includes ALTQ 2.1 code, which supports FreeBSD2, FreeBSD3,
1847 NetBSD and OpenBSD. For BSD/OS, ALTQ does not work.
1848 ALTQ in KAME supports (or tries to support) IPv6.
1849 (actually, ALTQ is developed on KAME repository since ALTQ 2.1 - Jan 2000)
1851 ALTQ occupies single character device number. For FreeBSD, it is officially
1852 allocated. For OpenBSD and NetBSD, we use the number which is not
1853 currently allocated (will eventually get an official number).
1854 The character device is enabled for i386 architecture only. To enable and
1855 compile ALTQ-ready kernel for other archititectures, take the following steps:
1856 - assume that your architecture is FOOBAA.
1857 - modify sys/arch/FOOBAA/FOOBAA/conf.c (or somewhere that defines cdevsw),
1858 to include a line for ALTQ. look at sys/arch/i386/i386/conf.c for
1859 example. The major number must be same as i386 case.
1860 - copy kernel configuration file (like ALTQ.v6 or GENERIC.v6) from i386,
1861 and modify accordingly.
1863 - before building userland, change netbsd/{lib,usr.sbin,usr.bin}/Makefile
1864 (or openbsd/foobaa) so that it will visit altq-related sub directories.
1868 6.1 KAME node as correspondent node
1870 Default installation recognizes home address option (in destination
1871 options header). No sub-options are supported. interaction with
1872 IPsec, and/or 2292bis API, needs further study.
1874 6.2 KAME node as home agent/mobile node
1876 KAME kit includes Ericsson mobile-ip6 code. The integration is just started
1877 (in Feb 2000), and we will need some more time to integrate it better.
1879 See kame/mip6config/{QUICKSTART,README_MIP6.txt} for more details.
1881 The Ericsson code implements revision 09 of the mobile-ip6 draft. There
1882 are other implementations available:
1883 NEC: http://www.6bone.nec.co.jp/mipv6/internal-dist/ (-13 draft)
1884 SFC: http://neo.sfc.wide.ad.jp/~mip6/ (-13 draft)
1888 The KAME developers basically do not make a bother about coding
1889 style. However, there is still some agreement on the style, in order
1890 to make the distributed develoment smooth.
1892 - the tab character should be 8 columns wide (tabstops are at 8, 16, 24, ...
1893 column). With vi, use ":set ts=8 sw=8".
1894 - each line should be within 80 characters.
1895 - keep a single open/close bracket in a comment such as in the following
1897 putchar('('); /* ) */
1898 without this, some vi users would have a hard time to match a pair of
1899 brackets. Although this type of bracket seems clumsy and is even
1900 harmful for some other type of vi users and Emacs users, the
1901 agreement in the KAME developers is to allow it.
1902 - add the following line to the head of every KAME-derived file:
1903 /* (dollar)KAME(dollar) */
1904 where "(dollar)" is the dollar character ($), and around "$" are tabs.
1905 (this is for C. For other language, you should use its own comment
1907 Once commited to the CVS repository, this line will contain its
1908 version number (see, for example, at the top of this file). This
1909 would make it easy to report a bug.
1910 - when creating a new file with the WIDE copyright, tap "make copyright.c" at
1911 the top-level, and use copyright.c as a template. KAME RCS tag will be
1912 included automatically.
1913 - when editting a third-party package, keep its own coding style as
1914 much as possible, even if the style does not follow the items above.
1916 When you want to contribute something to the KAME project, and if *you
1917 do not mind* the agreement, it would be helpful for the project to
1918 keep these rules. Note, however, that we would never intend to force
1919 you to adopt our rules. We would rather regard your own style,
1920 especially when you have a policy about the style.
1922 <end of IMPLEMENTATION>
projects / dragonfly.git / blob