idnits 2.17.1 draft-ietf-ipv6-router-selection-05.txt: -(572): Line appears to be too long, but this could be caused by non-ascii characters in UTF-8 encoding Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3667, Section 5.1 on line 13. -- Found old boilerplate from RFC 3978, Section 5.5 on line 640. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 651. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 658. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 664. ** Found boilerplate matching RFC 3978, Section 5.4, paragraph 1 (on line 632), which is fine, but *also* found old RFC 2026, Section 10.4C, paragraph 1 text on line 33. ** The document seems to lack an RFC 3978 Section 5.1 IPR Disclosure Acknowledgement -- however, there's a paragraph with a matching beginning. Boilerplate error? ** This document has an original RFC 3978 Section 5.4 Copyright Line, instead of the newer IETF Trust Copyright according to RFC 4748. ** This document has an original RFC 3978 Section 5.5 Disclaimer, instead of the newer disclaimer which includes the IETF Trust according to RFC 4748. ** The document uses RFC 3667 boilerplate or RFC 3978-like boilerplate instead of verbatim RFC 3978 boilerplate. After 6 May 2005, submission of drafts without verbatim RFC 3978 boilerplate is not accepted. The following non-3978 patterns matched text found in the document. That text should be removed or replaced: By submitting this Internet-Draft, I certify that any applicable patent or other IPR claims of which I am aware have been disclosed, or will be disclosed, and any of which I become aware will be disclosed, in accordance with RFC 3668. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- == There is 1 instance of lines with non-ascii characters in the document. == No 'Intended status' indicated for this document; assuming Proposed Standard Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** There are 3 instances of too long lines in the document, the longest one being 2 characters in excess of 72. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the RFC 3978 Section 5.4 Copyright Line does not match the current year -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (August 10, 2004) is 7197 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'RFC2893' is mentioned on line 88, but not defined ** Obsolete undefined reference: RFC 2893 (Obsoleted by RFC 4213) == Unused Reference: 'RFC2983' is defined on line 603, but no explicit reference was found in the text ** Obsolete normative reference: RFC 2461 (Obsoleted by RFC 4861) ** Obsolete normative reference: RFC 3775 (Obsoleted by RFC 6275) -- Obsolete informational reference (is this intentional?): RFC 2893 (ref. 'RFC2983') (Obsoleted by RFC 4213) Summary: 10 errors (**), 0 flaws (~~), 5 warnings (==), 8 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 IPng Working Group R. Draves 2 Internet Draft D. Thaler 3 Document: draft-ietf-ipv6-router-selection-05.txt Microsoft 4 August 10, 2004 6 Default Router Preferences and More-Specific Routes 8 Status of this Memo 10 By submitting this Internet-Draft, I certify that any applicable 11 patent or other IPR claims of which I am aware have been disclosed, 12 or will be disclosed, and any of which I become aware will be 13 disclosed, in accordance with RFC 3668. 15 Internet-Drafts are working documents of the Internet Engineering 16 Task Force (IETF), its areas, and its working groups. Note that 17 other groups may also distribute working documents as Internet- 18 Drafts. 20 Internet-Drafts are draft documents valid for a maximum of six 21 months and may be updated, replaced, or obsoleted by other documents 22 at any time. It is inappropriate to use Internet-Drafts as reference 23 material or to cite them other than as "work in progress." 25 The list of current Internet-Drafts can be accessed at 26 http://www.ietf.org/ietf/1id-abstracts.txt. 28 The list of Internet-Draft Shadow Directories can be accessed at 29 http://www.ietf.org/shadow.html. 31 Copyright Notice 33 Copyright (C) The Internet Society (2004). All Rights Reserved. 35 Abstract 37 This document describes an optional extension to Router 38 Advertisement messages for communicating default router preferences 39 and more-specific routes from routers to hosts. This improves the 40 ability of hosts to pick an appropriate router, especially when the 41 host is multi-homed and the routers are on different links. The 42 preference values and specific routes advertised to hosts require 43 administrative configuration; they are not automatically derived 44 from routing tables. 46 1. Introduction 48 Neighbor Discovery [RFC2461] specifies a conceptual model for hosts 49 that includes a Default Router List and a Prefix List. Hosts send 50 Router Solicitation messages and receive Router Advertisement 51 messages from routers. Hosts populate their Default Router List and 52 Prefix List based on information in the Router Advertisement 53 messages. A conceptual sending algorithm uses the Prefix List to 54 determine if a destination address is on-link and the Default Router 55 List to select a router for off-link destinations. 57 In some network topologies where the host has multiple routers on 58 its Default Router List, the choice of router for an off-link 59 destination is important. In some situations, one router may provide 60 much better performance than another for a destination. In other 61 situations, choosing the wrong router may result in a failure to 62 communicate. (A later section gives specific examples of these 63 scenarios.) 65 This document describes an optional extension to Neighbor Discovery 66 Router Advertisement messages for communicating default router 67 preferences and more-specific routes from routers to hosts. This 68 improves the ability of hosts to pick an appropriate router for an 69 off-link destination. 71 Neighbor Discovery provides a Redirect message that routers can use 72 to correct a host's choice of router. A router can send a Redirect 73 message to a host, telling it to use a different router for a 74 specific destination. However, the Redirect functionality is limited 75 to a single link. A router on one link cannot redirect a host to a 76 router on another link. Hence, Redirect messages do not help multi- 77 homed (through multiple interfaces) hosts select an appropriate 78 router. 80 Multi-homed hosts are an increasingly important scenario, especially 81 with IPv6. In addition to a wired network connection, like Ethernet, 82 hosts may have one or more wireless connections, like 802.11 or 83 Bluetooth. In addition to physical network connections, hosts may 84 have virtual or tunnel network connections. For example, in addition 85 to a direct connection to the public Internet, a host may have a 86 tunnel into a private corporate network. Some IPv6 transition 87 scenarios can add additional tunnels. For example, hosts may have 88 6to4 [RFC3056] or configured tunnel [RFC2893] network connections. 90 This document requires that the preference values and specific 91 routes advertised to hosts require explicit administrative 92 configuration. They are not automatically derived from routing 93 tables. In particular, the preference values are not routing metrics 94 and it is not recommended that routers "dump out" their entire 95 routing tables to hosts. 97 We use Router Advertisement messages, instead of some other protocol 98 like RIP [RFC2080], because Router Advertisements are an existing 99 standard, stable protocol for router-to-host communication. 100 Piggybacking this information on existing message traffic from 101 routers to hosts reduces network overhead. Neighbor Discovery shares 102 with Multicast Listener Discovery the property that they both define 103 host-to-router interactions, while shielding the host from having to 104 participate in more general router-to-router interactions. In 105 addition, RIP is unsuitable because it does not carry route 106 lifetimes so it requires frequent message traffic with greater 107 processing overheads. 109 The mechanisms specified here are backwards-compatible, so that 110 hosts that do not implement them continue to function as well as 111 they did previously. 113 1.1. Conventions used in this document 115 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 116 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in 117 this document are to be interpreted as described in [RFC2119]. 119 2. Message Formats 121 2.1. Preference Values 123 Default router preferences and preferences for more-specific routes 124 are encoded the same way. 126 Preference values are encoded in two bits, as follows: 127 01 High 128 00 Medium (default) 129 11 Low 130 10 Reserved - MUST NOT be sent 131 Note that implementations can treat the value as a two-bit signed 132 integer. 134 Having just three values reinforces that they are not metrics and 135 more values do not appear to be necessary for reasonable scenarios. 137 2.2. Changes to Router Advertisement Message Format 139 The changes from Neighbor Discovery [RFC2461] section 4.2 and 140 [RFC3775] section 7.1 are as follows: 142 0 1 2 3 143 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 144 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 145 | Type | Code | Checksum | 146 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 147 | Cur Hop Limit |M|O|H|Prf|Resvd| Router Lifetime | 148 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 149 | Reachable Time | 150 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 151 | Retrans Timer | 152 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 153 | Options ... 154 +-+-+-+-+-+-+-+-+-+-+-+- 156 Fields: 158 Prf (Default Router Preference) 159 2-bit signed integer. Indicates whether or not to prefer 160 this router over other default routers. If Router 161 Lifetime is zero, the preference value MUST be set to 162 (00) by the sender and MUST be ignored by the receiver. 163 If the Reserved (10) value is received, the receiver 164 MUST treat the value as if it were (00). 166 Resvd (Reserved) 167 A 3-bit unused field. It MUST be initialized to zero by 168 the sender and MUST be ignored by the receiver. 170 Possible Options: 172 Route Information 173 These options specify prefixes that are reachable via 174 the router. 176 Discussion: 178 Note that in addition to the preference value in the message header, 179 a Router Advertisement can also contain a Route Information Option 180 for ::/0, with a preference value and lifetime. Encoding a 181 preference value in the Router Advertisement header has some 182 advantages: 184 1. It allows for a distinction between the "best router for the 185 default route" and the "router least likely to redirect common 186 traffic", as described below in section 5.1. 188 2. When the best router for the default route is also the router 189 least likely to redirect common traffic (which will be a common 190 case), encoding the preference value in the message header is more 191 efficient than having to send a separate option. 193 2.3. Route Information Option 195 0 1 2 3 196 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 197 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 198 | Type | Length | Prefix Length |Resvd|Prf|Resvd| 199 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 200 | Route Lifetime | 201 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 202 | Prefix (Variable Length) | 203 . . 204 . . 205 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 207 Fields: 209 Type TBD 210 Length 8-bit unsigned integer. The length of the option 211 (including the Type and Length fields) in units of 212 8 octets. The Length field is 1, 2, or 3 depending on 213 Prefix Length. If Prefix Length is greater than 64, then 214 Length must be 3. If Prefix Length is greater than 0, 215 then Length must be 2 or 3. If Prefix Length is zero, 216 then Length must be 1, 2, or 3. 218 Prefix Length 219 8-bit unsigned integer. The number of leading bits in 220 the Prefix that are valid. The value ranges from 0 to 221 128. The Prefix field is 0, 8, or 16 octets depending on 222 Length. 224 Prf (Route Preference) 225 2-bit signed integer. The Route Preference indicates 226 whether to prefer the router associated with this prefix 227 over others, when multiple identical prefixes (for 228 different routers) have been received. If the Reserved 229 (10) value is received, the Route Information Option 230 MUST be ignored. 232 Resvd (Reserved) 233 Two 3-bit unused fields. They MUST be initialized to 234 zero by the sender and MUST be ignored by the receiver. 236 Route Lifetime 237 32-bit unsigned integer. The length of time in seconds 238 (relative to the time the packet is sent) that the 239 prefix is valid for route determination. A value of all 240 one bits (0xffffffff) represents infinity. 242 Prefix Variable-length field containing an IP address or a 243 prefix of an IP address. The Prefix Length field 244 contains the number of valid leading bits in the prefix. 245 The bits in the prefix after the prefix length (if any) 246 are reserved and MUST be initialized to zero by the 247 sender and ignored by the receiver. 249 Routers MUST NOT include two Route Information Options with the same 250 Prefix and Prefix Length in the same Router Advertisement. 252 Discussion: 254 There are several reasons for using a new Route Information Option, 255 instead of using flag bits to overload the existing Prefix 256 Information Option: 258 1. Prefixes will typically only show up in one or the other kind 259 of option, not both, so a new option does not introduce 260 duplication. 262 2. The Route Information Option is typically 16 octets while the 263 Prefix Information Option is 32 octets. 265 3. Using a new option may improve backwards-compatibility with 266 some host implementations. 268 3. Conceptual Model of a Host 270 There are three possible conceptual models for host implementation 271 of default router preferences and more-specific routes, 272 corresponding to different levels of support. We refer to these as 273 type A, type B, and type C. 275 3.1. Conceptual Data Structures for Hosts 277 Type A hosts ignore default router preferences and more-specific 278 routes. They use the conceptual data structures described in 279 Neighbor Discovery [RFC2461]. 281 Type B hosts use a Default Router List augmented with preference 282 values, but ignore all Route Information Options. They use the 283 Default Router Preference value in the Router Advertisement header. 284 They ignore Route Information Options. 286 Type C hosts use a Routing Table instead of a Default Router List. 287 (The Routing Table may also subsume the Prefix List, but that is 288 beyond the scope of this document.) Entries in the Routing Table 289 have a prefix, prefix length, preference value, lifetime, and next- 290 hop router. Type C hosts use both the Default Router Preference 291 value in the Router Advertisement header and Route Information 292 Options. 294 When a type C host receives a Router Advertisement, it modifies its 295 Routing Table as follows. When processing a Router Advertisement, a 296 type C host first updates a ::/0 route based on the Router Lifetime 297 and Default Router Preference in the Router Advertisement message 298 header. Then as the host processes Route Information Options in the 299 Router Advertisement message body, it updates its routing table for 300 each such option. The Router Preference and Lifetime values in a 301 ::/0 Route Information Option override the preference and lifetime 302 values in the Router Advertisement header. Updating each route is 303 done as follows. If the received route's lifetime is zero, the 304 route is removed from the Routing Table if present. If a route's 305 lifetime is non-zero, the route is added to the Routing Table if not 306 present and the route's lifetime and preference is updated if the 307 route is already present. A route is located in the Routing Table 308 based on prefix, prefix length, and next-hop router. 310 For example, suppose hosts receive a Router Advertisement from 311 router X with a Router Lifetime of 100 seconds and Default Router 312 Preference of Medium. The body of the Router Advertisement contains 313 a Route Information Option for ::/0 with a Route Lifetime of 200 314 seconds and a Route Preference of Low. After processing the Router 315 Advertisement, a type A host will have an entry for router X in its 316 Default Router List with lifetime 100 seconds. If a type B host 317 receives the same Router Advertisement, it will have an entry in its 318 Default Router List for router X with Medium preference and lifetime 319 100 seconds. A type C host will have an entry in its Routing Table 320 for ::/0 -> router X, with Low preference and lifetime 200 seconds. 321 A type C host MAY have a transient state, during processing of the 322 Router Advertisement, in which it has an entry in its Routing Table 323 for ::/0 -> router X with Medium preference and lifetime 100 324 seconds. 326 3.2. Conceptual Sending Algorithm for Hosts 328 Type A hosts use the conceptual sending algorithm described in 329 Neighbor Discovery [RFC2461]. 331 When a type B host does next-hop determination and consults its 332 Default Router List, it primarily prefers reachable routers over 333 non-reachable routers and secondarily uses the router preference 334 values. If the host has no information about the router's 335 reachability then the host assumes the router is reachable. 337 When a type C host does next-hop determination and consults its 338 Routing Table for an off-link destination, it first prefers 339 reachable routers over non-reachable routers, second uses longest- 340 matching-prefix, and third uses route preference values. Again, if 341 the host has no information about the router's reachability then the 342 host assumes the router is reachable. 344 If there are no routes matching the destination (i.e., no default 345 routes and no more-specific routes), then a type C host SHOULD 346 discard the packet and report a Destination Unreachable / No Route 347 To Destination error to the upper layer. 349 3.3. Destination Cache Management 351 When a type C host processes a Router Advertisement and updates its 352 conceptual Routing Table, it MUST invalidate or remove Destination 353 Cache Entries and redo next-hop determination for destinations 354 affected by the Routing Table changes. 356 3.4. Client Configurability 358 Type B and C hosts MAY be configurable with preference values that 359 override the values in Router Advertisements received. This is 360 especially useful for dealing with routers which may not support 361 preferences. 363 3.5. Router Reachability Probing 365 When a host avoids using any non-reachable router X and instead 366 sends a data packet to another router Y, and the host would have 367 used router X if router X were reachable, then the host SHOULD probe 368 each such router X's reachability by sending a single Neighbor 369 Solicitation to that router's address. A host MUST NOT probe a 370 router's reachability in the absence of useful traffic that the host 371 would have sent to the router if it were reachable. In any case, 372 these probes MUST be rate-limited to no more than one per minute per 373 router. 375 This requirement allows the host to discover when router X becomes 376 reachable and to start using router X at that point. Otherwise, the 377 host might not notice router X's reachability and continue to use 378 the less-desirable router Y until the next Router Advertisement is 379 sent by X. Note that the router may have been unreachable for 380 reasons other than being down (e.g., a switch in the middle being 381 down), so it may be up to 30 minutes (the maximum advertisement 382 period) before the next Router Advertisement would be sent. 384 For a type A host (following the algorithm in [RFC2461]), no probing 385 is needed since all routers are equally preferable. A type B or C 386 host, on the other hand, explicitly probes unreachable, preferable 387 routers to notice when they become reachable again. 389 3.6. Example 391 Suppose a type C host has four entries in its Routing Table: 392 ::/0 -> router W with Medium preference 393 2001::/16 -> router X with Medium preference 394 3ffe::/16 -> router Y with High preference 395 3ffe::/16 -> router Z with Low preference 396 and the host is sending to 3ffe::1, an off-link destination. If all 397 routers are reachable, then the host will choose router Y. If router 398 Y is not reachable, then router Z will be chosen and the 399 reachability of router Y will be probed. If routers Y and Z are not 400 reachable, then router W will be chosen and the reachability of 401 routers Y and Z will be probed. If routers W, Y, and Z are all not 402 reachable, then the host should use Y while probing the reachability 403 of W and Z. Router X will never be chosen because its prefix does 404 not match the destination. 406 4. Router Configuration 408 Routers should not advertise preferences or routes by default. In 409 particular, they should not "dump out" their entire routing table to 410 hosts. Routers MAY have a configuration mode where a filter is 411 applied to their routing table to obtain the routes that are 412 advertised to hosts. 414 Routers SHOULD NOT send more than 17 Route Information Options in 415 Router Advertisements per link. This arbitrary bound is meant to 416 reinforce that relatively few and carefully selected routes should 417 be advertised to hosts. 419 The preference values (both Default Router Preferences and Route 420 Preferences) should not be routing metrics or automatically derived 421 from metrics: the preference values should be configured. 423 The information contained in Router Advertisements may change 424 through actions of system management. For instance, the lifetime or 425 preference of advertised routes may change, new routes could be 426 added, etc. In such cases, the router MAY transmit up to 427 MAX_INITIAL_RTR_ADVERTISEMENTS unsolicited advertisements, using the 428 same rules as in [RFC2461]. When ceasing to be an advertising 429 interface and sending Router Advertisements with a Router Lifetime 430 of zero, the Router Advertisement SHOULD also set the Route Lifetime 431 to zero in all Route Information Options. 433 4.1. Guidance to Administrators 435 The High and Low (non-default) preference values should only be used 436 when someone with knowledge of both routers and the network topology 437 configures them explicitly. For example, it could be a common 438 network administrator, or it could be a customer request to 439 different administrators managing the routers. 441 As one exception to this general rule, the administrator of a router 442 that does not have a connection to the Internet, or is connected 443 through a firewall that blocks general traffic, should configure the 444 router to advertise a Low Default Router Preference. 446 In addition, the administrator of a router should configure the 447 router to advertise a specific route for the site prefix of the 448 network(s) to which the router belongs. The administrator may also 449 configure the router to advertise specific routes for directly 450 connected subnets and any shorter prefixes for networks to which the 451 router belongs. 453 For example, if a home user sets up a tunnel into a firewalled 454 corporate network, the access router on the corporate network end of 455 the tunnel should advertise itself as a default router, but with a 456 Low preference. Furthermore the corporate router should advertise a 457 specific route for the corporate site prefix. The net result is that 458 destinations in the corporate network will be reached via the 459 tunnel, and general Internet destinations will be reached via the 460 home ISP. Without these mechanisms, the home machine might choose to 461 send Internet traffic into the corporate network or corporate 462 traffic into the Internet, leading to communication failure because 463 of the firewall. 465 It is worth noting that the network administrator setting up 466 preferences and/or more specific routes in Routing Advertisements 467 typically does not know which kind of nodes (Type A, B, and/or C) 468 will be connected to its links. This requires that the administrator 469 will need to configure the settings that will work in an optimal 470 fashion no matter which kinds of nodes will be attached. 472 5. Examples 474 5.1. Best Router for ::/0 vs Router Least Likely to Redirect 476 The best router for the default route is the router with the best 477 route toward the wider Internet. The router least likely to 478 redirect traffic depends on the actual traffic usage. The two 479 concepts can be different when the majority of communication 480 actually needs to go through some other router. 482 For example, consider a situation where you have a link with two 483 routers X and Y. Router X is the best for 2002::/16. (It's your 6to4 484 site gateway.) Router Y is the best for ::/0. (It connects to the 485 native IPv6 Internet.) Router X forwards native IPv6 traffic to 486 router Y; router Y forwards 6to4 traffic to router X. If most 487 traffic from this site is sent to 2002:/16 destinations, then router 488 X is the one least likely to redirect. 490 To make type A hosts work well, both routers should advertise 491 themselves as default routers. In particular, if router Y goes down, 492 type A hosts should send traffic to router X to maintain 6to4 493 connectivity, so router X as well as router Y needs to be a default 494 router. 496 To make type B hosts work well, router X should in addition 497 advertise itself with a High default router preference. This will 498 cause type B hosts to prefer router X, minimizing the number of 499 redirects. 501 To make type C hosts work well, router X should in addition 502 advertise the ::/0 route with Low preference and the 2002::/16 route 503 with Medium preference. A type C host will end up with three routes 504 in its routing table: ::/0 -> router X (Low), ::/0 -> router Y 505 (Medium), 2002::/16 -> router X (Medium). It will send 6to4 traffic 506 to router X and other traffic to router Y. Type C hosts will not 507 cause any redirects. 509 Note that when type C hosts process the Router Advertisement from 510 router X, the Low preference for ::/0 overrides the High default 511 router preference. If the ::/0 specific route were not present, then 512 a type C host would apply the High default router preference to its 513 ::/0 route to router X. 515 5.2. Multi-Homed Host and Isolated Network 517 In another scenario, a multi-homed host is connected to the Internet 518 via router X on one link and to an isolated network via router Y on 519 another link. The multi-homed host might have a tunnel into a 520 firewalled corporate network, or it might be directly connected to 521 an isolated test network. 523 In this situation, a type A multi-homed host (which has no default 524 router preferences or more-specific routes) will have no way to 525 intelligently choose between the two routers X and Y on its Default 526 Router List. Users of the host will see unpredictable connectivity 527 failures, depending on the destination address and the choice of 528 router. 530 A multi-homed type C host in this same situation can correctly 531 choose between routers X and Y, if the routers are configured 532 appropriately. For example, router Y on the isolated network should 533 advertise a Route Information Option for the isolated network 534 prefix. It might not advertise itself as a default router at all 535 (zero Router Lifetime), or it might advertise itself as a default 536 router with Low preference. Router X should advertise itself as a 537 default router with Medium preference. 539 6. Security Considerations 541 A malicious node could send Router Advertisement messages, 542 specifying High Default Router Preference or carrying specific 543 routes, with the effect of pulling traffic away from legitimate 544 routers. However, a malicious node could easily achieve this same 545 effect in other ways. For example, it could fabricate Router 546 Advertisement messages with zero Router Lifetime from the other 547 routers, causing hosts to stop using the other routes. By 548 advertising a specific prefix, this attack could be carried out in a 549 less noticeable way. However, this attack has no significant 550 incremental impact on Internet infrastructure security. 552 A malicious node could also include an infinite lifetime in a Route 553 Information Option causing the route to linger indefinitely. A 554 similar attack already exists with Prefix Information Options in 555 RFC2461, where a malicious node causes a prefix to appear as on-link 556 indefinitely, resulting in lack of connectivity if it is not. In 557 contrast, an infinite lifetime in a Route Information Option will 558 cause router reachability probing to continue indefinitely, but will 559 not result in lack of connectivity. 561 [RFC3756] provides more details on the trust models, and there is 562 work in progress in the SEND WG on securing router discovery 563 messages that will address these problems. 565 7. IANA Considerations 567 Section 2.3 defines a new Neighbor Discovery [RFC2461] option, the 568 Route Information Option, which has been assigned the value TBD 569 within the numbering space for IPv6 Neighbor Discovery Option 570 Formats. 572 RFC EDITOR�s NOTE (to be removed prior to publication): the IANA is 573 requested to assign a value for "TBD" in the IPv6 Neighbor Discovery 574 Option Formats. When the assignment has been made, the RFC Editor 575 is asked to replace "TBD" (above in this section, and in section 576 2.3) with the assigned value and to remove this note. 578 8. Acknowledgments 580 The authors would like to acknowledge the contributions of Balash 581 Akbari, Steve Deering, Robert Elz, Tony Hain, Bob Hinden, Christian 582 Huitema, JINMEI Tatuya, Erik Nordmark, Pekka Savola, Kresimir 583 Segaric, and Brian Zill. The packet diagrams are derived from 584 Neighbor Discovery [RFC2461]. 586 9. Normative References 588 [RFC2461] Narten, T., Nordmark, E. and W. Simpson, "Neighbor 589 Discovery for IP Version 6 (IPv6)", RFC 2461, December 590 1998. 592 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 593 Requirement Levels", BCP 14, RFC 2119, March 1997. 595 [RFC3775] Johnson, D., Perkins, C. and J. Arkko, "Mobility Support 596 in IPv6", RFC 3775, June 2004. 598 10. Informative References 600 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 601 via IPv4 Clouds", RFC 3056, February 2001. 603 [RFC2983] Gilligan, R. and E. Nordmark, "Transition Mechanisms for 604 IPv6 Hosts and Routers", RFC 2893, August 2000. 606 [RFC2080] Malkin, G. and R. Minnear, "RIPng for IPv6", RFC 2080, 607 January 1997. 609 [RFC3756] Nikander, P., Ed., Kempf, J. and E. Nordmark, "IPv6 610 Neighbor Discovery (ND) Trust Models and Threats", RFC 611 3756, May 2004. 613 Authors' Addresses 615 Richard Draves 616 Microsoft Research 617 One Microsoft Way 618 Redmond, WA 98052 619 Phone: +1 425 706 2268 620 Email: richdr@microsoft.com 622 Dave Thaler 623 Microsoft 624 One Microsoft Way 625 Redmond, WA 98052 626 Phone: +1 425 703 8835 627 Email: dthaler@microsoft.com 628 Full Copyright Statement 630 Copyright (C) The Internet Society (2004). This document is subject 631 to the rights, licenses and restrictions contained in BCP 78, and 632 except as set forth therein, the authors retain all their rights. 634 This document and the information contained herein are provided on 635 an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE 636 REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE 637 INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR 638 IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 639 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 640 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 642 Intellectual Property 644 The IETF takes no position regarding the validity or scope of any 645 Intellectual Property Rights or other rights that might be claimed 646 to pertain to the implementation or use of the technology described 647 in this document or the extent to which any license under such 648 rights might or might not be available; nor does it represent that 649 it has made any independent effort to identify any such rights. 650 Information on the procedures with respect to rights in RFC 651 documents can be found in BCP 78 and BCP 79. 653 Copies of IPR disclosures made to the IETF Secretariat and any 654 assurances of licenses to be made available, or the result of an 655 attempt made to obtain a general license or permission for the use 656 of such proprietary rights by implementers or users of this 657 specification can be obtained from the IETF on-line IPR repository 658 at http://www.ietf.org/ipr. 660 The IETF invites any interested party to bring to its attention any 661 copyrights, patents or patent applications, or other proprietary 662 rights that may cover technology that may be required to implement 663 this standard. Please address the information to the IETF at ietf- 664 ipr@ietf.org.