idnits 2.17.1 draft-ietf-idmr-dvmrp-v3-09.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** Looks like you're using RFC 2026 boilerplate. This must be updated to follow RFC 3978/3979, as updated by RFC 4748. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- ** The document seems to lack a 1id_guidelines paragraph about 6 months document validity -- however, there's a paragraph with a matching beginning. Boilerplate error? ** The document is more than 15 pages and seems to lack a Table of Contents. == No 'Intended status' indicated for this document; assuming Proposed Standard == The page length should not exceed 58 lines per page, but there was 47 longer pages, the longest (page 2) being 59 lines == It seems as if not all pages are separated by form feeds - found 0 form feeds but 48 pages Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack separate sections for Informative/Normative References. All references will be assumed normative when checking for downward references. ** There is 1 instance of too long lines in the document, the longest one being 4 characters in excess of 72. ** The abstract seems to contain references ([Deer90], [Wait88]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. == There are 1 instance of lines with non-RFC6890-compliant IPv4 addresses in the document. If these are example addresses, they should be changed. == There are 1 instance of lines with multicast IPv4 addresses in the document. If these are generic example addresses, they should be changed to use the 233.252.0.x range defined in RFC 5771 ** The document seems to lack a both a reference to RFC 2119 and the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords. RFC 2119 keyword, line 48: '... The keywords MUST, MUST NOT, REQUIR...' RFC 2119 keyword, line 49: '... SHOULD NOT, RECOMMENDED, MAY, and O...' RFC 2119 keyword, line 134: '... MUST be set to 1....' RFC 2119 keyword, line 368: '...e. The checksum MUST be calculated up...' RFC 2119 keyword, line 369: '...transmission and MUST be validated on ...' (36 more instances...) == The 'Obsoletes: ' line in the draft header should list only the _numbers_ of the RFCs which will be obsoleted by this document (if approved); it should not include the word 'RFC' in the list. -- The draft header indicates that this document obsoletes RFC1075, but the abstract doesn't seem to directly say this. It does mention RFC1075 though, so this could be OK. Miscellaneous warnings: ---------------------------------------------------------------------------- == Line 326 has weird spacing: '...ets are encap...' == Line 359 has weird spacing: '...aft Ack for ...' == Line 1789 has weird spacing: '...for the purpo...' -- 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 (September 1999) is 8989 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: 'RFC-2119' is mentioned on line 51, but not defined -- Looks like a reference, but probably isn't: '00' on line 726 ** Downref: Normative reference to an Informational RFC: RFC 1071 (ref. 'Brad88') -- Possible downref: Non-RFC (?) normative reference: ref. 'Deer90' -- Possible downref: Non-RFC (?) normative reference: ref. 'Deer91' -- Possible downref: Non-RFC (?) normative reference: ref. 'Fen97b' ** Obsolete normative reference: RFC 1519 (ref. 'Full93') (Obsoleted by RFC 4632) ** Downref: Normative reference to an Informational RFC: RFC 1701 (ref. 'Han94a') ** Downref: Normative reference to an Informational RFC: RFC 1702 (ref. 'Han94b') -- Possible downref: Non-RFC (?) normative reference: ref. 'Ken98a' -- Possible downref: Non-RFC (?) normative reference: ref. 'Ken98b' -- Possible downref: Non-RFC (?) normative reference: ref. 'Ken98c' -- Possible downref: Non-RFC (?) normative reference: ref. 'McCl98' -- Possible downref: Non-RFC (?) normative reference: ref. 'Perl92' ** Downref: Normative reference to an Historic RFC: RFC 1518 (ref. 'Rekh93') -- Possible downref: Non-RFC (?) normative reference: ref. 'Reyn94' -- Possible downref: Non-RFC (?) normative reference: ref. 'Thal97' ** Downref: Normative reference to an Experimental RFC: RFC 1075 (ref. 'Wait88') Summary: 13 errors (**), 0 flaws (~~), 10 warnings (==), 14 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 T. Pusateri 2 INTERNET DRAFT Juniper Networks 3 Obsoletes: RFC 1075 September 1999 4 draft-ietf-idmr-dvmrp-v3-09.txt Expires: March 18, 2000 6 Distance Vector Multicast Routing Protocol 8 Status of this Memo 10 This document is an Internet-Draft and is in full conformance with 11 all provisions of Section 10 of RFC2026. 13 Internet-Drafts are working documents of the Internet Engineering 14 Task Force (IETF), its areas, and its working groups. Note that 15 other groups may also distribute working documents as Internet- 16 Drafts. 18 Internet-Drafts are draft documents valid for a maximum of six months 19 and may be updated, replaced, or obsoleted by other documents at any 20 time. It is inappropriate to use Internet- Drafts as reference 21 material or to cite them other than as "work in progress." 23 The list of current Internet-Drafts can be accessed at 24 http://www.ietf.org/ietf/1id-abstracts.txt 26 The list of Internet-Draft Shadow Directories can be accessed at 27 http://www.ietf.org/shadow.html. 29 Abstract 31 DVMRP is an Internet routing protocol that provides an efficient 32 mechanism for connection-less datagram delivery to a group of hosts 33 across an internetwork. It is a distributed protocol that dynamically 34 generates IP Multicast delivery trees using a technique called 35 Reverse Path Multicasting (RPM) [Deer90]. This document is an update 36 to Version 1 of the protocol specified in RFC 1075 [Wait88]. 38 1. Introduction 40 DVMRP uses a distance vector distributed routing algorithm in order 41 to build per-source-group multicast delivery trees. A good 42 introduction to distance vector routing can be found in [Perl92]. 43 The application of distance vector routing to multicast tree 44 formulation is described in [Deer91]. 46 1.1. Requirements Terminology 48 The keywords MUST, MUST NOT, REQUIRED, SHALL, SHALL NOT, SHOULD, 49 SHOULD NOT, RECOMMENDED, MAY, and OPTIONAL, when they appear in this 50 document, are to be interpreted as described in [RFC-2119]. 52 1.2. Reverse Path Multicasting 54 Datagrams follow multicast delivery trees from a source to all 55 members of a multicast group [Deer89], replicating the packet only at 56 necessary branches in the delivery tree. The trees are calculated and 57 updated dynamically to track the membership of individual groups. 58 When a datagram arrives on an interface, the reverse path to the 59 source of the datagram is determined by examining a DVMRP routing 60 table of known source networks. If the datagram arrives on an 61 interface that would be used to transmit datagrams back to the 62 source, then it is forwarded to the appropriate list of downstream 63 interfaces. Otherwise, it is not on the optimal delivery tree and 64 should be discarded. In this way duplicate packets can be filtered 65 when loops exist in the network topology. The source specific 66 delivery trees are automatically pruned back as group membership 67 changes or routers determine that no group members are present. This 68 keeps the delivery trees to the minimum branches necessary to reach 69 all of the group members. New sections of the tree can also be added 70 dynamically as new members join the multicast group by grafting the 71 new sections onto the delivery trees. 73 1.3. Tunnel Encapsulation 75 Because not all IP routers support native multicast routing, DVMRP 76 includes direct support for tunneling IP Multicast datagrams through 77 routers. The IP Multicast datagrams are encapsulated in unicast IP 78 packets and addressed to the routers that do support native multicast 79 routing. DVMRP treats tunnel interfaces in an identical manner to 80 physical network interfaces. 82 In previous implementations, DVMRP protocol messages were sent un- 83 encapsulated to the unicast tunnel endpoint address. While this was 84 more direct, it increased the complexity of firewall configuration. 85 The most noticeable change in this specification regarding tunnels is 86 that all DVMRP protocol messages should be sent encapsulated across 87 the tunnel. Previously, protocol messages were sent un-encapsulated 88 directly to the tunnel endpoint. See Appendix C for backward 89 compatibility issues. 91 Note: All protocol messages sent on point-to-point links (including 92 tunnels) should use a destination address of All-DVMRP-Routers. This 93 change will allow the protocol messages to be forwarded across 94 multicast-only tunnels without making encapsulation and decapsulation 95 difficult. 97 In practice, tunnels typically use either IP-IP [Perk96] or Generic 98 Routing Encapsulation (GRE) [Han94a,Han94b], although, other 99 encapsulation methods are acceptable. 101 1.4. Document Overview 103 Section 2 provides an overview of the protocol and the different 104 message types exchanged by DVMRP routers. Those who wish to gain a 105 general understanding of the protocol but are not interested in the 106 more precise details may wish to only read this section. Section 3 107 explains the detailed operation of the protocol to accommodate 108 developers needing to provide inter-operable implementations. 109 Included in Appendix A, is a summary of the DVMRP parameters. A 110 section on DVMRP support for tracing and troubleshooting is the topic 111 of Appendix B. Finally, a short DVMRP version compatibility section 112 is provided in Appendix C to assist with backward compatibility 113 issues. 115 2. Protocol Overview 117 DVMRP can be summarized as a "broadcast & prune" multicast routing 118 protocol. It builds per-source broadcast trees based upon routing 119 exchanges, then dynamically creates per-source-group multicast 120 delivery trees by pruning (removing branches from) the source's 121 truncated broadcast tree. It performs Reverse Path Forwarding checks 122 to determine when multicast traffic should be forwarded to downstream 123 interfaces. In this way, source-rooted shortest path trees can be 124 formed to reach all group members from each source network of 125 multicast traffic. 127 2.1. Neighbor Discovery 129 Neighbor DVMRP routers are discovered dynamically by sending Neighbor 130 Probe Messages on local multicast capable network interfaces and 131 tunnel pseudo interfaces. These messages are sent periodically to the 132 All-DVMRP-Routers [Reyn94] IP Multicast group address. (See Appendix 133 C for backwards compatibility issues.) The IP TTL of these messages 134 MUST be set to 1. 136 Each Neighbor Probe message contains the list of Neighbor DVMRP 137 routers for which Neighbor Probe messages have been received on that 138 interface. In this way, Neighbor DVMRP routers can ensure that they 139 are seen by each other. 141 Once you have received a Probe from a neighbor that contains your 142 address in the neighbor list, you have established a two-way neighbor 143 adjacency with this router. 145 2.2. Source Location 147 When an IP Multicast datagram is received by a router running DVMRP, 148 it first looks up the source network in the DVMRP routing table. The 149 interface on which the best route to the source of the datagram was 150 received is called the upstream (also called RPF) interface. If the 151 datagram arrived on the correct upstream interface, then it is a 152 candidate for forwarding to one or more downstream interfaces. If the 153 datagram did not arrive on the anticipated upstream interface, it is 154 discarded. This check is known as a reverse path forwarding check and 155 must be performed by all DVMRP routers. 157 In order to ensure that all DVMRP routers have a consistent view of 158 the path back to a source, a routing table is propagated to all DVMRP 159 routers as an integral part of the protocol. Each router advertises 160 the network number and mask of the interfaces it is directly 161 connected to as well as relaying the routes received from neighbor 162 routers. DVMRP requires an interface metric to be configured on all 163 physical and tunnel interfaces. When a route is received, the metric 164 of the interface over which the datagram was received must be added 165 to the metric of the route being advertised in the route report 166 message. This adjusted metric should be used when comparing metrics 167 to determine the best upstream neighbor. 169 Although there is certainly additional overhead associated with 170 propagating a separate DVMRP routing table, it does provide two nice 171 features. First, since all DVMRP routers are exchanging the same 172 routes, there are no inconsistencies between routers when determining 173 the upstream interface (aside from normal convergence issues related 174 to distance vector routing protocols). By placing the burden of 175 synchronization on the protocol as opposed to the network manager, 176 DVMRP reduces the risk of creating routing loops or black holes due 177 to disagreement between neighbor routers on the upstream interface. 179 Second, by propagating its own routing table, DVMRP makes it 180 convenient to have separate paths for unicast versus multicast 181 datagrams. Although, ideally, many network managers would prefer to 182 keep their unicast and multicast traffic aligned, tunneled multicast 183 topologies may prevent this causing the unicast and multicast paths 184 to diverge. Additionally, service providers may prefer to keep the 185 unicast and multicast traffic separate for routing policy reasons as 186 they experiment with IP multicast routing and begin to offer it as a 187 service. 189 2.3. Dependent Downstream Routers 191 In addition to providing a consistent view of source networks, the 192 exchange of routes in DVMRP provides one other important feature. 193 DVMRP uses the route exchange as a mechanism for upstream routers to 194 determine if any downstream routers depend on them for forwarding 195 from particular source networks. DVMRP accomplishes this by using a 196 technique called "Poison Reverse". If a downstream router selects an 197 upstream router as the best next hop to a particular source network, 198 this is indicated by echoing back the route on the upstream interface 199 with a metric equal to the original metric plus infinity. When the 200 upstream router receives the report and sees a metric that lies 201 between infinity and twice infinity, it can then add the downstream 202 router from which it received the report to a list of dependent 203 routers for this source. 205 This list of dependent routers per source network built by the 206 "Poison Reverse" technique will provide the foundation necessary to 207 determine when it is appropriate to prune back the IP source specific 208 multicast trees. 210 2.4. Designated Forwarder 212 When two or more multicast routers are connected to a multi-access 213 network, it could be possible for duplicate packets to be forwarded 214 on the network (one copy from each router). DVMRP prevents this 215 possibility by electing a forwarder for each source as a side effect 216 of its route exchange. When two routers on a multi-access network 217 exchange source networks, each of the routers will know the others 218 metric back to each source network. Therefore, of all the DVMRP 219 routers on a shared network, the router with the lowest metric to a 220 source network is responsible for forwarding data on to the shared 221 network. If two or more routers have an equally low metric, the 222 router with the lowest IP address becomes the designated forwarder 223 for the network. In this way, DVMRP does an implicit designated 224 forwarder election for each source network on each downstream 225 interface. 227 2.5. Building Multicast Trees 229 As previously mentioned, when an IP multicast datagram arrives, the 230 upstream interface is determined by looking up the interface on which 231 the best route to the source of the datagram was received. If the 232 upstream interface is correct, then a DVMRP router will forward the 233 datagram to a list of downstream interfaces. 235 2.5.1. Adding Local Group Members 237 The IGMP local group database is maintained by all IP multicast 238 routers on each physical, multicast capable network [Fen97a]. If the 239 destination group address is listed in the local group database, and 240 the router is the designated forwarder for the source, then the 241 interface is included in the list of downstream interfaces. If there 242 are no group members on the interface, then the interface is removed 243 from the outgoing interface list. 245 2.5.2. Adding Interfaces with Neighbors 247 Initially, all interfaces with downstream dependent neighbors should 248 be included in the downstream interface list when a forwarding cache 249 entry is first created. This allows the downstream routers to be 250 aware of traffic destined for a particular (source network, group) 251 pair. The downstream routers will then have the option to send prunes 252 and subsequent grafts for this (source network, group) pair as 253 requirements change from their respective downstream routers and 254 local group members. 256 2.6. Pruning Multicast Trees 258 As mentioned above, routers at the edges will remove their interfaces 259 that have no group members associated with an IP multicast datagram. 260 If a router removes all of its downstream interfaces, it notifies the 261 upstream router that it no longer wants traffic destined for a 262 particular (source network, group) pair. This is accomplished by 263 sending a DVMRP Prune message upstream to the router it expects to 264 forward datagrams from a particular source. 266 Recall that a downstream router will inform an upstream router that 267 it depends on the upstream router to receive datagrams from 268 particular source networks by using the "Poison Reverse" technique 269 during the exchange of DVMRP routes. This method allows the upstream 270 router to build a list of downstream routers on each interface that 271 are dependent upon it for datagrams from a particular source network. 272 If the upstream router receives prune messages from each one of the 273 dependent downstream routers on an interface, then the upstream 274 router can in turn remove this interface from its downstream 275 interface list. If the upstream router is able to remove all of its 276 downstream interfaces in this way, it can then send a DVMRP Prune 277 message to its upstream router. This continues until the unneeded 278 branches are removed from the delivery tree. 280 In order to remove old prune state information for (source network, 281 group) pairs that are no longer active, it is necessary to limit the 282 life of a prune and periodically resume the broadcasting procedure. 283 The prune message contains a prune lifetime, indicating the length of 284 time that the prune should remain in effect. When the prune lifetime 285 expires, the interface is joined back onto the multicast delivery 286 tree. If unwanted multicast datagrams continue to arrive, the prune 287 mechanism will be re-initiated and the cycle will continue. If all 288 of the downstream interfaces are removed from a multicast delivery 289 tree causing a DVMRP Prune message to be sent upstream, the lifetime 290 of the prune sent must be equal to the minimum of the remaining 291 lifetimes of the received prunes. 293 2.7. Grafting Multicast Trees 295 Once a tree branch has been pruned from a multicast delivery tree, 296 packets from the corresponding (source network, group) pair will no 297 longer be forwarded. However, since IP multicast supports dynamic 298 group membership, hosts may join a multicast group at any time. In 299 this case, DVMRP routers use Grafts to cancel the prunes that are in 300 place from the host back on to the multicast delivery tree. A router 301 will send a Graft message to its upstream neighbor if a group join 302 occurs for a group that the router has previously sent a prune. 303 Separate Graft messages must be sent to the appropriate upstream 304 neighbor for each source network that has been pruned. Since there 305 would be no way to tell if a Graft message sent upstream was lost or 306 the source simply quit sending traffic, it is necessary to 307 acknowledge each Graft message with a DVMRP Graft Ack message. If an 308 acknowledgment is not received within a Graft Time-out period, the 309 Graft message should be retransmitted using binary exponential back- 310 off between retransmissions. Duplicate Graft Ack messages should 311 simply be ignored. The purpose of the Graft Ack message is to simply 312 acknowledge the receipt of a Graft message. It does not imply that 313 any action was taken as a result of receiving the Graft message. 314 Therefore, all Graft messages received from a neighbor with whom a 315 two-way neighbor relationship has been formed should be acknowledged 316 whether or not they cause an action on the receiving router. 318 3. Detailed Protocol Operation 320 This section contains a detailed description of DVMRP. It covers 321 sending and receiving of DVMRP messages as well as the generation and 322 maintenance of IP Multicast forwarding cache entries. 324 3.1. Protocol Header 326 DVMRP packets are encapsulated in IP datagrams, with an IP protocol 327 number of 2 (IGMP) as specified in the Assigned Numbers RFC [Reyn94]. 328 All fields are transmitted in Network Byte Order. DVMRP packets use a 329 common protocol header that specifies the IGMP [Fen97a] Packet Type 330 as hexadecimal 0x13 (DVMRP). DVMRP protocol packets should be sent 331 with the Precedence field in the IP header set to Internetwork 332 Control (hexadecimal 0xc0 for the Type of Service Octet) [Post81]. A 333 diagram of the common protocol header follows: 335 0 8 16 31 336 +---------+---------+--------------------+ 337 | Type | Code | Checksum | 338 |(0x13) | | | 339 +---------+---------+----------+---------+ 340 | Reserved | Minor | Major | 341 | | Version |Version | 342 +-------------------+----------+---------+ 344 Figure 1 - Common Protocol Header 346 A Major Version of 3 and a Minor Version of 0xFF should be used to 347 indicate compliance with this specification. The value of the Code 348 field determines the DVMRP packet type. Currently, there are codes 349 allocated for DVMRP protocol message types as well as protocol 350 analysis and troubleshooting packets. The protocol message Codes 351 are: 353 Code Packet Type Description 354 ---------------------------------------------------------------- 355 1 DVMRP Probe for neighbor discovery 356 2 DVMRP Report for route exchange 357 7 DVMRP Prune for pruning multicast delivery trees 358 8 DVMRP Graft for grafting multicast delivery trees 359 9 DVMRP Graft Ack for acknowledging graft messages 360 ---------------------------------------------------------------- 362 Table 1 - Standard Protocol Packet Types 364 There are additional codes used for protocol analysis and 365 troubleshooting. These codes are discussed in Appendix B. 367 The Checksum is the 16-bit one's complement of the one's complement 368 sum of the DVMRP message. The checksum MUST be calculated upon 369 transmission and MUST be validated on reception of a packet. The 370 checksum of the DVMRP message is calculated with the checksum field 371 set to zero. See [Brad88] for more information. 373 3.2. Probe Messages 375 When a DVMRP router is configured to run on an interface (physical or 376 tunnel), it multicasts DVMRP Probe packets to inform other DVMRP 377 routers that it is operational. Effectively, they serve three 378 purposes. 380 1. Probes provide a mechanism for DVMRP routers to locate each other. 381 DVMRP sends on each interface, a Probe Message containing the list 382 of the neighbors detected for that specific interface. If no 383 DVMRP neighbors are found, the network is considered to be a leaf 384 network. 386 2. Probes provide a way for DVMRP routers to determine the 387 capabilities of each other. This may be deduced from the major and 388 minor version numbers in the Probe packet or directly from the 389 capability flags. These flags were first introduced to allow 390 optional protocol features. This specification now mandates the 391 use of Generation Id's and pruning and, therefore, provides no 392 optional capabilities. Other capability flags were used for 393 tracing and troubleshooting and are no longer a part of the actual 394 protocol. 396 3. Probes provide a keep-alive function in order to quickly detect 397 neighbor loss. Probes sent on each multicast capable interface 398 configured for DVMRP SHOULD use an interval of 10 seconds. The 399 neighbor time-out interval SHOULD be set at 35 seconds. This 400 allows fairly early detection of a lost neighbor yet provides 401 tolerance for busy multicast routers. These values MUST be 402 coordinated between all DVMRP routers on a physical network 403 segment. 405 3.2.1. Router Capabilities 407 In the past, there have been many versions of DVMRP in use with a 408 wide range of capabilities. Practical considerations require a 409 current implementation to inter-operate with these older 410 implementations that don't formally specify their capabilities and 411 are not compliant with this specification. For instance, for major 412 versions less than 3, it can be assumed that the neighbor does not 413 support pruning. The formal capability flags were first introduced 414 in an well known implementation (Mrouted version 3.5) in an attempt 415 to take the guess work out which features are supported by a 416 neighbor. Many of these flags are no longer necessary since they are 417 now a required part of the protocol, however, special consideration 418 is necessary to not confuse older implementations that expect these 419 flags to be set. Appendix C was written to assist with these and 420 other backward compatibility issues. 422 Three of the flags were used for actual protocol operation. The 423 other two assigned flags were used for troubleshooting purposes which 424 should be documented in a separate specification. All of the bits 425 marked "U" in the Figure below are now unused. They may be defined in 426 the future and MUST be set to 0 on transmission and ignored on 427 reception. Bit position 0 is the LEAF bit which is a current research 428 topic. It MUST be set to 0 and ignored on reception. Bit positions 429 1, 2, and 3 MUST be set to 1 for backward compatibility. They were 430 used to specify the PRUNE, GENID, and MTRACE bits. The first two, 431 PRUNE and GENID, are now required features. The MTRACE bit must be 432 set so existing implementations will not assume this neighbor does 433 not support multicast trace-route [Fen97b]. However, since this bit 434 is now reserved and set to 1, newer implementations should not use 435 this bit in the Probe message to determine if multicast trace-route 436 is supported by a neighbor. Instead, the M bit should only be used in 437 a Neighbors2 message as described in Appendix B. The bit marked S 438 stands for SNMP capable. This bit is used by troubleshooting 439 applications and should only be tested in the Neighbors2 message. 441 The N bit (which stands for Netmask) is defined by this 442 specification. It is used to indicate the neighbor will accept 443 network masks appended to the Prune, Graft, and Graft Ack messages. 444 This bit only indicates that the neighbor understands the netmask. It 445 DOES NOT mean that Prune, Graft, and Graft Ack messages sent to this 446 neighbor must include a netmask. Refer to the sections on Prune, 447 Graft, and Graft Ack messages for more details. 449 7 6 5 4 3 2 1 0 450 U U N S M G P L 451 +---+---+---+---+----+---+---+---+ 452 |0 |0 |X | 0 | 1 |1 |1 | 0 | 453 +---+---+---+---+----+---+---+---+ 455 Figure 2 - Probe Capability Flags 457 3.2.2. Generation ID 459 If a DVMRP router is restarted, it will not be aware of any previous 460 prunes that it had sent or received. In order for the neighbor to 461 detect that the router has restarted, a non-decreasing number is 462 placed in the periodic probe message called the generation ID. When 463 a change in the generation ID is detected, any prune information 464 received from the router is no longer valid and should be flushed. 465 If this prune state has caused prune information to be sent upstream, 466 a graft will need to be sent upstream just as though a new member has 467 joined below. Once data begins to be delivered downstream, if the 468 downstream router again decides to be pruned from the delivery tree, 469 a new prune can be sent upstream at that time. 471 In addition, the effects of a restart can be minimized if the router 472 can learn all of the routes known by its neighbors without having to 473 wait for an entire report interval to pass. When a router detects a 474 change in the generation ID of a neighbor, it should send a unicast 475 copy of its entire routing table to the neighbor. 477 A time of day clock provides a good source for a non-decreasing 32 478 bit integer. 480 3.2.3. Neighbor Addresses 482 As a DVMRP router sees Probe messages from its DVMRP neighbors, it 483 records the neighbor addresses on each interface and places them in 484 the Probe message sent on the particular interface. This allows the 485 neighbor router to know that its probes have been received by the 486 sending router. 488 In order to minimize one-way neighbor relationships, a router MUST 489 delay sending poison route reports in response to routes advertised 490 by a neighbor until the neighbor includes the routers address in its 491 probe messages. On point-to-point interfaces and tunnel pseudo- 492 interfaces, this means that no packets should be forwarded onto these 493 interfaces until two-way neighbor relationships have formed. 495 Implementations written before this specification will not wait 496 before sending reports nor will they ignore reports sent. Therefore, 497 reports from these implementations SHOULD be accepted whether or not 498 a probe with the routers address has been received. 500 3.2.4. Neighbor Time-Out 502 When a neighbor times out, the following steps should be taken: 504 1. All routes learned from this neighbor should be immediately placed 505 in hold-down. All downstream dependencies ON this neighbor should 506 be removed. 508 2. If this neighbor is considered to be the designated forwarder for 509 any of the routes it is advertising, a new designated forwarder 510 for each source network should be selected. 512 3. Any forwarding cache entries based on this upstream neighbor 513 should be flushed. 515 4. Any outstanding Grafts awaiting acknowledgments from this router 516 should be flushed. 518 5. All downstream dependencies received FROM this neighbor should be 519 removed. Forwarding cache entries should be checked to see if 520 this is the last downstream dependent neighbor on the interface. 521 If so, and this router isn't the designated forwarder (with local 522 group members present), the interface should be removed. 524 It is possible as an optimization to send a prune upstream if this 525 causes the last downstream interface to be removed. However, this 526 prune could be unnecessary if no more traffic is arriving. It is 527 also acceptable to simply wait for traffic to arrive before 528 sending the prune upstream. 530 3.2.5. Probe Packet Format 532 The Probe packet is variable in length depending on the number of 533 neighbor IP addresses included. The length of the IP packet can be 534 used to determine the number of neighbors in the Probe message. The 535 current Major Version is 3. 537 7 15 23 31 538 +---------+--------------+--------------------+ 539 | Type | Code | Checksum | 540 | (0x13) | (0x1) | | 541 +---------+--------------+----------+---------+ 542 | | | | | 543 |Reserved | Capabilities | Minor | Major | 544 +---------+--------------+----------+---------+ 545 | | 546 | Generation ID | 547 +---------------------------------------------+ 548 | | 549 | Neighbor IP Address 1 | 550 +---------------------------------------------+ 551 | | 552 | Neighbor IP Address 2 | 553 +---------------------------------------------+ 554 | | 555 | ... | 556 +---------------------------------------------+ 557 | | 558 | Neighbor IP Address N | 559 +---------------------------------------------+ 561 Figure 3 - DVMRP Probe Packet Format 563 Generation ID 564 This field contains a non-decreasing number used to detect a 565 change in neighbor state. 567 Neighbor IP Address N 568 This is a list of Neighbor IP addresses whom the sending router 569 has received Probe messages from. 571 3.2.6. IGMP Designated Querier Election 573 Since it is wasteful to have more than a single router sending IGMP 574 Host Membership Queries on a given physical network, a single router 575 on each physical network is elected as the Designated Querier. This 576 election was formerly a part of DVMRP. However, this is now 577 specified as a part of the IGMP version 2 protocol. See Appendix C 578 for details on backwards compatibility. 580 Even though only one router will act as the IGMP designated querier, 581 all DVMRP routers must use IGMP to learn local group memberships. 583 3.3. Multicast Forwarding 585 DVMRP was originally designed to forward multicast packets by 586 building the downstream interface list for each packet as it arrived. 587 However, to reduce per packet processing time, the result of the 588 first lookup MAY be cached in a forwarding table. Then, as routes, 589 downstream dependent neighbors, or group membership change, the cache 590 forwarding table entries MUST be updated to reflect these changes. 592 3.3.1. Designated Forwarder 594 Initially, a DVMRP router should assume it is the designated 595 forwarder for all source networks on all downstream interfaces. As it 596 receives route reports, it can determine if other routers on multi- 597 access networks have better routes back to a particular source 598 network. A route is considered better if the adjusted received metric 599 is less than the metric that it will advertise for the source network 600 on the received interface or if the metrics are the same but the IP 601 address of the neighbor is lower. 603 If this neighbor becomes unreachable or starts advertising a worse 604 metric, then the router should become the designated forwarder for 605 this source network again on the downstream interface until it hears 606 from a better candidate. 608 If the upstream RPF interface changes, then the router should become 609 the designated forwarder on the previous upstream interface (which is 610 now a potential downstream interface) until it hears from a better 611 candidate. 613 3.3.2. Determining the upstream interface 615 When a multicast packet arrives, a DVMRP router will use the DVMRP 616 routing table to determine which interface leads back to the source. 617 If the packet did not arrive on that interface, it MUST be discarded 618 without further processing. Each multicast forwarding entry should 619 cache the upstream interface for a particular source host or source 620 network after looking this up in the DVMRP routing table. 622 3.3.3. Determining the downstream interface list 624 The downstream interface list is built by starting with the list of 625 non-leaf interfaces. The upstream interface MUST be removed from this 626 list. Then any interfaces on the list where all of the downstream 627 dependents have sent prunes upstream MUST be removed. Next, any 628 interfaces for which the router is the designated forwarder and local 629 group members are present MUST be added to the list. 631 3.4. Route Exchange 633 The routing information propagated by DVMRP is used for determining 634 the reverse path neighbor back to the source of the multicast 635 traffic. The interface used to reach this neighbor is called the 636 upstream interface. Tunnel pseudo-interfaces are considered to be 637 distinct from the physical interface on which the packet is actually 638 transmitted for the purpose of determining upstream and downstream 639 interfaces. 641 The routing information that is propagated by DVMRP contains a list 642 of source networks and an appropriate metric. The metric used is a 643 hop count which is incremented by the cost of the incoming interface 644 metric. Traditionally, physical interfaces use a metric of 1 while 645 the metric of a tunnel interface varies with the distance and 646 bandwidth in the path between the two tunnel endpoints. Users are 647 encouraged to configure tunnels with the same metric in each 648 direction to create symmetric routing and provide for easier problem 649 determination although the protocol does not strictly enforce this. 651 3.4.1. Source Network Aggregation 653 Implementations may wish to provide a mechanism to aggregate source 654 networks to reduce the size of the routing table. All implementations 655 should be able to accept reports for aggregated source networks in 656 accordance with Classless Inter-Domain Routing (CIDR) as described in 657 [Rekh93] and [Full93]. 659 There are two places where aggregation is particularly useful. 661 1. At organizational boundaries to limit the number of source 662 networks advertised out of the organization. 664 2. Within an organization to summarize non-local routing information 665 by using a default (0/0) route. 667 If an implementation wishes to support source aggregation, it MUST 668 transmit Prune and Graft messages according to the following rules: 670 A. If a Prune is received on a downstream interface for which the 671 source network advertised to that neighbor is an aggregate, then 672 if a prune is sent upstream, it should only be sent for the 673 contributing route based on the source address in the received 674 prune. 676 If additional data is received for sources within the range of the 677 aggregate, then this SHOULD trigger additional prunes to be sent 678 upstream for these sources. 680 There may be active forwarding cache entries for other 681 contributing routes to the aggregate. Prunes should not be sent 682 upstream to the contributing routes that have no forwarding state. 684 B. If a Graft is received on a downstream interface for which the 685 source network advertised to that neighbor is an aggregate 686 generated by the receiving router, then Graft messages MUST be 687 sent upstream (if necessary) for each route that contributed to 688 the aggregate that had been previously pruned. 690 3.4.2. Route Packing and Ordering 692 Since DVMRP Route Reports may need to refresh several thousand routes 693 each report interval, routers MUST attempt to spread the routes 694 reported across the whole route update interval. This reduces the 695 chance of synchronized route reports causing routers to become 696 overwhelmed for a few seconds each report interval. Since the route 697 report interval is 60 seconds, it is suggested that the total number 698 routes being updated be split across multiple Route Reports sent at 699 regular intervals. There was an earlier requirement that Route 700 Reports MUST contain source network/mask pairs sorted first by 701 increasing network mask and then by increasing source network. This 702 restriction has been lifted. Implementations conforming to this 703 specification MUST be able to receive Route Reports containing any 704 mixture of network masks and source networks. 706 In order to pack more source networks into a route report, source 707 networks are often represented by less than 4 octets. The number of 708 non-zero bytes in the mask value is used to determine the number of 709 octets used to represent each source network within that particular 710 mask value. For instance if the mask value of 255.255.0.0 is being 711 reported, the source networks would only contain 2 octets each. DVMRP 712 assumes that source networks will never be aggregated into networks 713 whose prefix length is less than 8. Therefore, it does not carry the 714 first octet of the mask in the Route Report since, given this 715 assumption, the first octet will always be 0xFF. This means that the 716 netmask value will always be represented in 3 octets. This method of 717 specifying source network masks is compatible with techniques 718 described in [Rekh93] and [Full93] to group traditional Class C 719 networks into super-nets and to allow different subnets of the same 720 Class A network to be discontinuous. It does not, however, allow 721 grouping class A networks into super-nets since the first octet of 722 the netmask is always assumed to be 255. 724 In this notation, the default route is represented as the least three 725 significant octets of the netmask [00 00 00], followed by one octet 726 for the network number [00]. This special case MUST be interpreted 727 as 0.0.0.0/0.0.0.0 and NOT 0.0.0.0/255.0.0.0. 729 3.4.3. Route Metrics 731 For each source network reported, a route metric is associated with 732 the route being reported. The metric is the sum of the interface 733 metrics between the router originating the report and the source 734 network. For the purposes of DVMRP, the Infinity metric is defined to 735 be 32. This limits the breadth across the whole DVMRP network and is 736 necessary to place an upper bound on the convergence time of the 737 protocol. 739 As seen in the packet format below, Route Reports do not contain a 740 count of the number of routes reported for each netmask. Instead, a 741 "Last" bit is defined as the high order bit of the octet following 742 the network address. This bit is set to signify when the last route 743 is being reported for a particular mask value. When the "Last" bit 744 is set and the end of the message has not been reached, the next 745 value will be a new netmask to be applied to the subsequent list of 746 routes. 748 3.4.4. Route Dependencies 750 In order for pruning to work correctly, each DVMRP router needs to 751 know which downstream routers depend on it for receiving datagrams 752 from particular source networks. Initially, when a new datagram 753 arrives from a particular source/group pair, it is broadcasted to all 754 downstream interfaces that have DVMRP neighbors who have indicated a 755 dependency on the receiving DVMRP router for that particular source. 756 A downstream interface can only be removed when the router has 757 received Prune messages from each of the dependent routers on that 758 interface. Each downstream router uses Poison Reverse to indicate 759 for which source networks it is dependent upon the upstream router. 760 The downstream router indicates this by echoing back the source 761 networks it expects to receive from the upstream router with infinity 762 added to the advertised metric. This means that the legal values for 763 the metric now become between 1 and (2 x Infinity - 1) or 1 and 63. 764 Values between 1 and 31 indicate reachable source networks. The value 765 Infinity (32) indicates the source network is not reachable. Values 766 between 33 and 63 indicate that the downstream router originating the 767 Report is depending upon the upstream router to provide multicast 768 datagrams from the corresponding source network. 770 3.4.5. Sending Route Reports 772 All of the active routes MUST be advertised over all interfaces with 773 neighbors present each Route Report Interval. In addition, flash 774 updates MAY be sent as needed but flash updates MUST NOT happen more 775 often than the Minimum Flash Update Interval (5 seconds). Flash 776 updates reduce the chances of routing loops and black holes occurring 777 when source networks become unreachable through a particular path. 778 Flash updates need only contain the source networks that have 779 changed. 781 When a router sees its own address in a neighbor probe packet for the 782 first time, it should send a unicast copy of its entire routing table 783 to the neighbor to reduce start-up time. 785 Reports should not be sent to a neighbor until a router has seen its 786 own address in the neighbors Probe router list. See Appendix C for 787 exceptions. 789 3.4.6. Receiving Route Reports 791 After receiving a route report, a check should be made to verify it 792 is from a known neighbor. Two-way neighbor relationships are 793 essential for proper DVMRP operation. Therefore, route reports from 794 unknown neighbors MUST be discarded. 796 In the following discussion, "Metric" refers to the metric of the 797 route as received in the route report. "Adjusted Metric" refers to 798 the metric of the route after the incoming interface metric has been 799 added. 801 If the metric received is less than infinity but the Adjusted Metric 802 is greater than or equal to infinity, the Adjusted Metric should be 803 set to infinity. 805 If the metric is greater than or equal to infinity, then no 806 adjustment of the metric should be made. 808 Each route in the report is then parsed and processed according to 809 the following rules: 811 A. If the route is new and the Adjusted Metric is less than infinity, 812 the route should be added. 814 B. If the route already exists, several checks must be performed. 816 1. Received Metric < infinity 818 If the neighbor was considered a downstream dependent neighbor, 819 the dependency is canceled. 821 In the following cases, the designated forwarder on one of the 822 downstream interfaces should be updated: 824 - If the Metric received would cause the router to advertise a 825 better metric on a downstream interface than the existing 826 designated forwarder for the source network on that 827 interface (or advertised metric would be the same but the 828 router's IP address is lower than the existing designated 829 forwarder on that interface). Then the receiving router 830 becomes the new designated forwarder for that source network 831 on that interface. If this router had sent a prune upstream 832 that is still active, it will need to send a graft. 834 - If the metric being advertised by the current designated 835 forwarder is worse than the receiving routers metric that it 836 would advertise on the receiving interface (from learning 837 the same route from a neighbor on another interface) or the 838 metric is the same but the receiving router has a lower IP 839 address, then the receiving router becomes the new 840 designated forwarder on that interface. This may trigger a 841 graft to be sent upstream. 843 - If the metric received for the source network is better than 844 the metric of the existing designated forwarder, save the 845 new designated forwarder and the metric it is advertising. 846 It is necessary to maintain knowledge of the current 847 designated forwarder for each source network in case the 848 time-out value for this neighbor is reached. If the time-out 849 is reached, then the designated forwarder responsibility for 850 the source network should be assumed. 852 A route is considered better when either the received Metric is 853 lower than the existing metric or the received Metric is the 854 same but the advertising router's IP address is lower. 856 a. Adjusted Metric > existing metric 858 If the Adjusted Metric is greater than the existing metric, 859 then check to see if the same neighbor is reporting the 860 route. If so, update the route metric and schedule a flash 861 update containing the route. Otherwise, skip to the next 862 route in the report. 864 b. Adjusted Metric < existing metric 866 Update the metric for the route and if the neighbor 867 reporting the route is different, update the upstream 868 neighbor in the routing table. Schedule a flash update 869 containing the route to downstream neighbors and a flash 870 poison update containing the route should be sent upstream 871 indicating a change in downstream dependency (even if its on 872 the same upstream interface). 874 c. Adjusted metric = existing metric 876 If the neighbor reporting the route is the same as the 877 existing upstream neighbor, then simply refresh the route. 878 If the neighbor is the same and the route is in hold-down, 879 it is permissible to prematurely take the route out of hold- 880 down and begin advertising it with a non-infinity metric. 881 If the route is taken out of hold-down, a flash update 882 containing the route should be scheduled on all DVMRP 883 interfaces except the one over which it was received. 885 If the neighbor reporting the route has a lower IP address 886 than the existing upstream neighbor, then switch to this 887 neighbor as the best route. Schedule a flash update 888 containing the route to downstream neighbors and a flash 889 poison update containing the route should be sent upstream 890 indicating a change in downstream dependency (even if its on 891 the same upstream interface). 893 2. Received Metric = infinity 894 If the neighbor was considered to be the designated 895 forwarder, the receiving router should now become the 896 designated forwarder for the source network on this 897 interface. 899 a. Next hop = existing next hop 901 If the existing metric was less than infinity, the route is 902 now unreachable. Delete the route and schedule a flash 903 update containing the route to all interfaces for which you 904 are the designated forwarder or have downstream dependents. 906 b. Next hop != existing next hop 908 The route can be ignored since the existing next hop has a 909 metric better than or equal to this next hop. 911 If the neighbor was considered a downstream dependent 912 neighbor, this should be canceled. 914 3. infinity < Received Metric < 2 x infinity 916 The neighbor considers the receiving router to be upstream for 917 the route and is indicating it is dependent on the receiving 918 router. 920 If the neighbor was considered to be the designated forwarder, 921 the receiving router should now become the designated forwarder 922 for the source network on this interface. 924 a. Neighbor on downstream interface 926 If the sending neighbor is considered to be on a downstream 927 interface for that route then the neighbor is to be 928 registered as a downstream dependent router for that route. 930 If this is the first time the neighbor has indicated 931 downstream dependence for this source and one or more prunes 932 have been sent upstream containing this source network, then 933 Graft messages MUST be sent upstream in the direction of the 934 source network for each group with existing prune state. 936 b. Neighbor on upstream interface 938 If the receiving router thinks the neighbor is on the 939 upstream interface, then the route should be treated as if 940 an infinity metric was received (See Received Metric = 941 infinity). 943 4. 2 x infinity <= Received Metric 945 If the Received Metric is greater than or equal to 2 x 946 infinity, the Metric is illegal and the route should be 947 ignored. 949 3.4.7. Route Expiration 951 A route expires if it has not been refreshed within the Route 952 Expiration time. This should be set to 2 x Route Report Interval + 20 953 (or 140) seconds. Due to flash updates, routes will typically not 954 expire but instead be removed in response to receiving an infinity 955 metric for the route. However, since not all existing 956 implementations implement flash updates, route expiration is 957 necessary. 959 3.4.8. Route Hold-down 961 When a route is deleted (because it expires, the neighbor it was 962 learned from goes away, or an infinity metric is received for the 963 route) a router may be able to reach the source network described by 964 the route through an alternate gateway. However, in the presence of 965 complex topologies, often, the alternate gateway may only be echoing 966 back the same route learned via a different path. If this occurs, the 967 route will continue to be propagated long after it is no longer 968 valid. 970 In order to prevent this, it is common in distance vector protocols 971 to continue to advertise a route that has been deleted with a metric 972 of infinity for one or more report intervals. This is called Hold- 973 down. A route MUST only be advertised with an infinity metric while 974 it is in hold-down. The hold-down period is 2 Report Intervals. 976 When a route goes into hold-down, all forwarding cache entries based 977 on the route should be flushed. 979 During the hold-down period, the route may be learned via a different 980 gateway or the same gateway with a different metric. The router MAY 981 use this new route for delivering to local group members. Also, 982 installing a new route during the hold-down period allows the route 983 to be advertised with a non-infinity metric more quickly once the 984 hold-down period is over. 986 In order to minimize outages caused by flapping routes, it is 987 permissible to prematurely take a route out of hold-down ONLY if the 988 route is re-learned from the SAME router with the SAME metric. 990 Route hold-down is not effective if only some routers implement it. 991 Therefore, it is now a REQUIRED part of the protocol. 993 3.4.9. Graceful Shutdown 995 During a graceful shutdown, an implementation MAY want to inform 996 neighbor routers that it is terminating. Routes that have been 997 advertised with a metric less than infinity should now be advertised 998 with a metric equal to infinity. This will allow neighbor routers to 999 switch more quickly to an alternate path for a source network if one 1000 exists. 1002 Routes that have been advertised with a metric between infinity and 2 1003 x infinity (indicating downstream neighbor dependence) should now be 1004 advertised with a metric equal to infinity (canceling the downstream 1005 dependence). 1007 3.4.10. Route Report Packet Format 1009 The format of a sample Route Report Packet is shown in Figure 4 1010 below. The packet shown is an example of how the source networks are 1011 packed into a Report. The number of octets in each Source Network 1012 will vary depending on the mask value. The values below are only an 1013 example for clarity and are not intended to represent the format of 1014 every Route Report. 1016 7 15 23 31 1017 +-----------+------------+-------------------------+ 1018 | Type | Code | Checksum | 1019 | (0x13) | (0x2) | | 1020 +-----------+------------+------------+------------+ 1021 | Reserved | Minor | Major | 1022 | | Version | Version | 1023 +-----------+------------+------------+------------+ 1024 | Mask1 | Mask1 | Mask1 | Src | 1025 | Octet2 | Octet3 | Octet4 | Net11 | 1026 +-----------+------------+------------+------------+ 1027 | SrcNet11(cont.)... | Metric11 | Src | 1028 | | | Net12 | 1029 +------------------------+------------+------------+ 1030 | SrcNet12(cont.)... | Metric12 | Mask2 | 1031 | | | Octet2 | 1032 +-----------+------------+------------+------------+ 1033 | Mask2 | Mask2 | SrcNet21 | 1034 | Octet3 | Octet4 | | 1035 +-----------+------------+------------+------------+ 1036 | SrcNet21(cont.)... | Metric21 | Mask3 | 1037 | | | Octet2 | 1038 +-----------+------------+------------+------------+ 1039 | Mask3 | Mask3 | ... | 1040 | Octet3 | Octet4 | | 1041 +-----------+------------+-------------------------+ 1043 Figure 4 - Example Route Report Packet Format 1045 3.5. Pruning 1047 DVMRP is described as a broadcast and prune multicast routing 1048 protocol since datagrams are initially sent out all dependent 1049 downstream interfaces forming a tree rooted at the source of the 1050 data. As the routers at the leaves of the tree begin to receive 1051 unwanted multicast traffic, they send prune messages upstream toward 1052 the source. This results in the multicast tree for a given source 1053 network and a given set of receivers. 1055 3.5.1. Leaf Networks 1057 Detection of leaf networks is very important to the pruning process. 1058 Routers at the end of a source specific multicast delivery tree must 1059 detect that there are no further downstream dependent routers. This 1060 detection mechanism is covered above in section 3.2 titled Probe 1061 Messages. If there are no group members present for a particular 1062 multicast datagram received, the leaf routers will start the pruning 1063 process by removing their downstream interfaces and sending a prune 1064 to the upstream router for that source. 1066 3.5.2. Source Networks 1068 By default, prunes are meant to be applied to a group and source 1069 network. However, it is possible to include a Netmask in the Prune 1070 message to alter this behavior. If no Netmask is included, a prune 1071 sent upstream triggered by traffic received from a particular source 1072 applies to all sources on that network. If a Netmask is included, it 1073 MUST first be validated. If the Netmask is a host mask, only that 1074 source address should be pruned. Otherwise, the Netmask MUST match 1075 the mask sent to the downstream router for that source. If it does 1076 not match the mask that the upstream router expected, the upstream 1077 router MUST ignore the prune and should log an error. When a 1078 aggregate source network is advertised downstream, the Netmask in the 1079 prune will match the mask of the aggregate route that was advertised. 1081 If the Prune message only contains the host address of the source 1082 (and not the corresponding Netmask), the source network can be 1083 determined easily by a best-match lookup using the routing table 1084 distributed as a part of DVMRP. 1086 3.5.3. Receiving a Prune 1088 When a prune is received, the following steps should be taken: 1090 1. If the neighbor is unknown, discard the received prune. 1092 2. Ensure the prune message contains at least the correct amount of 1093 data. 1095 3. Copy the source address, group address, and prune time-out value. 1096 If it is available in the packet, copy the Netmask value. 1097 Determine route to which prune applies. 1099 4. If there is no active source information for the (source network, 1100 group) pair, then ignore the prune. 1102 5. Verify that the prune was received from a dependent neighbor for 1103 the source network. If not, discard the prune. 1105 6. Determine if a prune is currently active from the same dependent 1106 neighbor for this (source network, group) pair. 1108 7. If so, reset the timer to the new time-out value. Otherwise, 1109 create state for the new prune and set a timer for the prune 1110 lifetime. 1112 8. Determine if all dependent downstream routers on the interface 1113 from which the prune was received have now sent prunes. 1115 9. If so, then determine if there are group members active on the 1116 interface and if this router is the designated forwarder for the 1117 network. 1119 10. If not, then remove the interface from all forwarding cache 1120 entries for this group instantiated using the route to which the 1121 prune applies. 1123 3.5.4. Sending a Prune 1125 When a forwarding cache is being used, there is a trade-off that should 1126 be considered when deciding when Prune messages should be sent upstream. 1127 In all cases, when a data packet arrives and the downstream interface 1128 list is empty, a prune is sent upstream. However, when a forwarding 1129 cache entry transistions to an empty downstream interface list it is 1130 possible as an optimization to send a prune at this time as well. This 1131 prune will possibly stop unwanted traffic sooner at the expense of 1132 sending extra prune traffic for sources that are no longer sending. 1133 When sending a prune upstream, the following steps should be taken: 1135 1. Decide if upstream neighbor is capable of receiving prunes. 1137 2. If not, then proceed no further. 1139 3. Stop any pending Grafts awaiting acknowledgments. 1141 4. Determine the prune lifetime. This value should be the minimum of 1142 the default prune lifetime (randomized to prevent synchronization) 1143 and the remaining prune lifetimes of the downstream neighbors. 1145 5. Form and transmit the packet to the upstream neighbor for the 1146 source. 1148 3.5.5. Retransmitting a Prune 1150 By increasing the prune lifetime to ~2 hours, the effect of a lost 1151 prune message becomes more apparent. Therefore, an implementation 1152 SHOULD retransmit prunes messages using binary exponential back-off 1153 during the lifetime of the prune if traffic is still arriving on the 1154 upstream interface. 1156 One way to implement this would be to send a prune, install a 1157 negative cache entry for 3 seconds while waiting for the prune to 1158 take effect. Then remove the negative cache entry. If traffic 1159 continues to arrive, a new forwarding cache request will be 1160 generated. The prune can be resent with the remaining prune lifetime 1161 and a negative cache entry can be installed for 6 seconds. After 1162 this, the negative cache entry is removed. This procedure is repeated 1163 while each time doubling the length of time the negative cache entry 1164 is installed. 1166 In addition to using binary exponential back-off, the interval 1167 between subsequent retransmissions should also be randomized to 1168 prevent synchronization. 1170 On multi-access networks, even if a prune is received by the upstream 1171 router, data may still be received due to other receivers (i.e. group 1172 members or other downstream dependent routers) on the network. 1174 3.5.6. Prune Packet Format 1176 A Prune Packet contains three required fields: the source host IP 1177 address, the destination group IP address, and the Prune Lifetime in 1178 seconds. An optional source network mask may also be appended to the 1179 Prune message. This mask applied to the Source Host Address will 1180 indicate the route that the prune applies to. A Source Network Mask 1181 field should only be sent in a Prune message to a neighbor if that 1182 neighbor has advertised the ability to process it by setting the 1183 Netmask capabilities bit. The length of the Prune message will 1184 indicate if the Source Network Mask has been included or not. 1186 The Prune Lifetime is a derived value calculated as the minimum of 1187 the default prune lifetime (2 hours) and the remaining lifetimes of 1188 any downstream prunes received for this source network and group. A 1189 router with no downstream dependent neighbors would use the the 1190 default prune lifetime (randomized to prevent synchronization). 1192 7 15 23 31 1193 +-----------+------------+-------------------------+ 1194 | Type | Code | Checksum | 1195 | (0x13) | (0x7) | | 1196 +-----------+------------+------------+------------+ 1197 | Reserved | Minor | Major | 1198 +------------------------+------------+------------+ 1199 | Source Host Address | 1200 +--------------------------------------------------+ 1201 | Group Address | 1202 +--------------------------------------------------+ 1203 | Prune Lifetime | 1204 +--------------------------------------------------+ 1205 | Source Network Mask | 1206 +--------------------------------------------------+ 1208 Figure 5 - Prune Packet Format 1210 Source Host Address 1211 The source host IP address of the datagram that triggered the 1212 prune. 1214 Group Address 1215 The destination group IP address of the datagram that triggered 1216 the prune. 1218 Prune Lifetime 1219 The number of seconds for which the upstream neighbor should keep 1220 this prune active. 1222 Source Network Mask 1223 The (optional) netmask of the route this prune applies to. 1225 3.6. Grafting 1227 Once a multicast delivery tree has been pruned back, DVMRP Graft 1228 messages are necessary to join new receivers onto the multicast tree. 1229 Graft messages are sent upstream hop-by-hop from the new receiver's 1230 first-hop router until a point on the multicast tree is reached. 1231 Since there is no way to tell whether a graft message was lost or the 1232 source stopped sending, each Graft message is acknowledged hop by 1233 hop. This ensures that the Graft message is not lost somewhere along 1234 the path between the receiver's first-hop router and the closest 1235 point on the multicast delivery tree. 1237 One or more Graft messages should be sent under the following 1238 conditions: 1240 1. A new local member joins a group that has been pruned upstream and 1241 this router is the designated forwarder for the source. 1243 2. A new dependent downstream router appears on a pruned branch. 1245 3. A dependent downstream router on a pruned branch restarts (new 1246 Generation ID). 1248 4. A Graft Retransmission Timer expires before a Graft-Ack is 1249 received. 1251 3.6.1. Sending a Graft 1253 Recall that by default, Prunes are source network specific and are 1254 sent up different trees for each source network. Grafts are sent in 1255 response to various conditions which are not necessarily source 1256 specific. Therefore, it may be necessary to send separate Graft 1257 messages to the appropriate upstream routers to counteract each 1258 previous source network specific prune that was sent. 1260 As mentioned above, a Graft message sent to the upstream DVMRP router 1261 should be acknowledged hop by hop guaranteeing end-to-end delivery. 1262 In order to send a Graft message, the following steps should be 1263 taken: 1265 1. Verify a prune exists for the source network and group. 1267 2. Verify that the upstream router is capable of receiving prunes 1268 (and therefore grafts). 1270 3. Add the graft to the retransmission timer list awaiting an 1271 acknowledgment. 1273 4. Formulate and transmit the Graft packet. 1275 If a Graft Acknowledgment is not received within the Graft 1276 Retransmission Time-out period, the Graft should be resent to the 1277 upstream router. The initial retransmission period is 5 seconds. A 1278 binary exponential back-off policy is used on subsequent 1279 retransmissions. 1281 3.6.2. Receiving a Graft 1283 1. If the neighbor is unknown, discard the received graft. 1285 2. Ensure the graft message contains at least the correct amount of 1286 data. 1288 3. Send back a Graft Ack to the sender. 1290 4. If the sender was a downstream dependent neighbor from which a 1291 prune had previously been received, then remove the prune state 1292 for this neighbor. If necessary, any forwarding cache entries 1293 based on this (source, group) pair should be updated to include 1294 this downstream interface. 1296 5. If a prune had been sent upstream, this may trigger a graft to 1297 now be sent to the upstream router. 1299 3.6.3. Graft Packet Format 1301 The format of a Graft packet is show below: 1303 7 15 23 31 1304 +-----------+------------+-------------------------+ 1305 | Type | Code | Checksum | 1306 | (0x13) | (0x8) | | 1307 +-----------+------------+------------+------------+ 1308 | Reserved | Minor | Major | 1309 +------------------------+------------+------------+ 1310 | Source Host Address | 1311 +--------------------------------------------------+ 1312 | Group Address | 1313 +--------------------------------------------------+ 1314 | Source Network Mask | 1315 +--------------------------------------------------+ 1317 Figure 6 - Graft Packet Format 1319 Source Host Address 1320 The source host IP address indicating which source host or source 1321 network to Graft. 1323 Group Address 1324 The destination group IP address to be grafted. 1326 Source Network Mask 1327 The (optional) netmask of the route this graft applies to. 1329 3.6.4. Sending a Graft Acknowledgment 1331 A Graft Acknowledgment packet is sent to a downstream neighbor in 1332 response to receiving a Graft message. All Graft messages MUST be 1333 acknowledged. This is true even if no other action is taken in 1334 response to receiving the Graft to prevent the source from 1335 continually re-transmitting the Graft message. The Graft 1336 Acknowledgment packet is identical to the Graft packet except that 1337 the DVMRP code in the common header is set to Graft Ack. This allows 1338 the receiver of the Graft Ack message to correctly identify which 1339 Graft was acknowledged and stop the appropriate retransmission timer. 1341 3.6.5. Receiving a Graft Acknowledgment 1343 When a Graft Acknowledgment is received, ensure the message contains 1344 at least the correct amount of data. The (source address, group) 1345 pair in the packet can be used to determine if a Graft was sent to 1346 this particular upstream router. If no Graft was sent, the Graft Ack 1347 can simply be ignored. If a Graft was sent, and the acknowledgment 1348 has come from the correct upstream router, then it has been 1349 successfully received and the retransmission timer for the Graft can 1350 be stopped. 1352 3.6.6. Graft Acknowledgment Packet Format 1354 The format of a Graft Ack packet (which is identical to that of a 1355 Graft packet) is show below: 1357 7 15 23 31 1358 +-----------+------------+-------------------------+ 1359 | Type | Code | Checksum | 1360 | (0x13) | (0x9) | | 1361 +-----------+------------+------------+------------+ 1362 | Reserved | Minor | Major | 1363 +------------------------+------------+------------+ 1364 | Source Host Address | 1365 +--------------------------------------------------+ 1366 | Group Address | 1367 +--------------------------------------------------+ 1368 | Source Network Mask | 1369 +--------------------------------------------------+ 1371 Figure 7 - Graft Ack Packet Format 1373 Source Host Address 1374 The source host IP address that was received in the Graft message. 1376 Group Address 1377 The destination group IP address that was received in the Graft 1378 message. 1380 Source Network Mask 1381 The (optional) netmask of the route this Graft Ack applies to. 1383 3.7. Interfaces 1385 Interfaces running DVMRP will either be multicast capable physical 1386 interfaces or encapsulated tunnel pseudo-interfaces. Physical 1387 interfaces may either be multi-access networks or point-to-point 1388 networks. Tunnel interfaces are used when there are non-multicast 1389 capable routers between DVMRP neighbors. Protocol messages and 1390 multicast data traffic are sent between tunnel endpoints using a 1391 standard encapsulation method [Perk96,Han94a,Han94b]. The unicast IP 1392 addresses of the tunnel endpoints are used as the source and 1393 destination IP addresses in the outer IP header. The inner IP header 1394 remains unchanged from the original packet. 1396 Protocol messages on point-to-point links should always use a 1397 destination IP address of All-DVMRP-Routers for ALL message types. 1398 While Prune, Graft, and Graft-Ack messages are only intended for a 1399 single recipient, the use of a multicast destination address is 1400 necessary for un-numbered links and encapsulated interfaces. 1402 When multiple addresses are configured on a single interface, it is 1403 necessary that all routers on the interface know about the same set 1404 of network addresses. In this way, each router will make the same 1405 choice for the designated forwarder for each source. In addition, a 1406 router configured with multiple addresses on an interface should 1407 consistently use the same address when sending DVMRP control 1408 messages. 1410 The maximum packet length of any DVMRP message should be the maximum 1411 packet size required to be forwarded without fragmenting. The use of 1412 Path MTU Discovery [Mogu90] is encouraged to determine this size. In 1413 the absence of Path MTU, the Requirements for Internet Hosts [Brad89] 1414 specifies this number as 576 octets. Be sure to consider the size of 1415 the encapsulated IP header as well when calculating the maximum size 1416 of a DVMRP protocol message. 1418 4. IANA Considerations 1420 The Internet Assigned Numbers Authority (IANA) is the central 1421 coordinator for the assignment of unique parameter values for 1422 Internet protocols. DVMRP uses IGMP [Fen97a] IP protocol messages to 1423 communicate between routers. The IGMP Type field is hexadecimal 0x13. 1425 On IP multicast capable networks, DVMRP uses the All-DVMRP-Routers 1426 local multicast group. This group address is 224.0.0.4. 1428 5. Network Management Considerations 1430 DVMRP provides several methods for network management monitoring and 1431 troubleshooting. Appendix B describes a request/response mechanism to 1432 directly query DVMRP neighbor information. In addition, a Management 1433 Information Base for DVMRP is defined in [Thal97]. 1435 A Management Information Base for the multicast forwarding cache is 1436 defined in [McCl98]. 1438 Also, a protocol independent multicast trace-route facility is 1439 defined in [Fen97b]. 1441 6. Security Considerations 1443 Security for DVMRP follows the general security architecture provided 1444 for the Internet Protocol [Ken98a]. This framework provides for both 1445 privacy and authentication. It recommends the use of the IP 1446 Authentication Header [Ken98b] to provide trusted neighbor 1447 relationships. Confidentiality is provided by the addition of the IP 1448 Encapsulating Security Payload [Ken98c]. 1450 7. References 1452 [Brad88] Braden, R., Borman, D., Partridge, C., "Computing the 1453 Internet Checksum", RFC 1071, September 1988. 1455 [Brad89] Braden, R., "Requirements for Internet Hosts -- 1456 Communication Layers", RFC 1122, October 1989. 1458 [Deer89] Deering, S., "Host Extensions for IP Multicasting", RFC 1459 1112, August 1989. 1461 [Deer90] Deering, S., Cheriton, D., "Multicast Routing in Datagram 1462 Internetworks and Extended LANs", ACM Transactions on 1463 Computer Systems, Vol. 8, No. 2, May 1990, pp. 85-110. 1465 [Deer91] Deering, S., "Multicast Routing in a Datagram 1466 Internetwork", PhD thesis, Electric Engineering Dept., 1467 Stanford University, December 1991. 1469 [Fen97a] Fenner, W., "Internet Group Management Protocol, Version 1470 2", RFC 2236, November 1997. 1472 [Fen97b] Fenner, W., Casner, S., "A "traceroute" facility for IP 1473 Multicast", Work In Progress, November 1997. 1475 [Full93] Fuller, V., T. Li, J. Yu, and K. Varadhan, "Classless 1476 Inter-Domain Routing (CIDR): an Address Assignment and 1477 Aggregation Strategy", RFC 1519, September 1993. 1479 [Han94a] Hanks, S., Li, T, Farinacci, D., and P. Traina, "Generic 1480 Routing Encapsulation", RFC 1701, NetSmiths, Ltd., and 1481 cisco Systems, October 1994. 1483 [Han94b] Hanks, S., Li, T., Farinacci, D., and P. Traina, "Generic 1484 Routing Encapsulation over IPv4 networks", RFC 1702, 1485 NetSmiths, Ltd., cisco Systems, October 1994. 1487 [Ken98a] Kent, S., Atkinson, R. "Security Architecture for the 1488 Internet Protocol", Work in Progress, February 1998. 1490 [Ken98b] Kent, S., Atkinson, R., "IP Authentication Header", Work in 1491 Progress, February 1998. 1493 [Ken98c] Kent, S., Atkinson, R., "IP Encapsulating Security Payload 1494 (ESP)", Work in Progress, February 1998. 1496 [McCl98] McCloghrie, K., Farinacci, D., Thaler, D., "IP Multicast 1497 Routing MIB", Work In Progress, July 1998. 1499 [Mogu90] Mogul, J., Deering, S., "Path MTU Discovery", RFC 1191, 1500 November 1990. 1502 [Perk96] Perkins, C., "IP Encapsulation within IP", RFC 2003, 1503 October 1996. 1505 [Perl92] Perlman, R., "Interconnections: Bridges and Routers", 1506 Addison-Wesley, May 1992, pp. 205-211. 1508 [Post81] Postel, J., "Internet Protocol", RFC 791, September, 1981. 1510 [Rekh93] Rekhter, Y., and T. Li, "An Architecture for IP Address 1511 Allocation with CIDR", RFC 1518, September 1993. 1513 [Reyn94] Reynolds, J., Postel, J., "Assigned Numbers", STD 0002, 1514 October 1994. 1516 [Thal97] Thaler, D., "Distance-Vector Multicast Routing Protocol 1517 MIB", Work In Progress, April 1997. 1519 [Wait88] Waitzman, D., Partridge, C., Deering, S., "Distance Vector 1520 Multicast Routing Protocol", RFC 1075, November 1988. 1522 8. Author's Address 1524 Thomas Pusateri 1525 Juniper Networks, Inc. 1526 385 Ravendale Dr. 1527 Mountain View, CA 94043 1528 Phone: (650) 526-8046 1529 EMail: pusateri@juniper.net 1531 9. Acknowledgments 1533 The author would like to acknowledge the original designers of the 1534 protocol, Steve Deering, Craig Partridge, and David Waitzman. 1535 Version 3 of the protocol would not have been possible without the 1536 original work of Ajit Thyagarajan and the ongoing (and seemingly 1537 endless) work of Bill Fenner. Credit also goes to Danny Mitzel and 1538 Dave Thaler for the careful review of this document and Nitin Jain, 1539 Dave LeRoy, Charles Mumford, Ravi Shekhar, and Shuching Shieh for 1540 their helpful comments. 1542 10. Appendix A - Constants & Configurable Parameters 1544 The following table provides a summary of the DVMRP timing 1545 parameters: 1547 Parameter Value (seconds) 1548 ---------------------------------------------------------- 1549 Probe Interval 10 1550 Neighbor Time-out Interval 35 1551 Minimum Flash Update Interval 5 1552 Route Report Interval 60 1553 Route Expiration Time 140 1554 Route Hold-down 2 x Route Report Interval 1555 Prune Lifetime variable (< 2 hours) 1556 Prune Retransmission Time 3 with exp. back-off 1557 Graft Retransmission Time 5 with exp. back-off 1558 ---------------------------------------------------------- 1560 Table 2 - Parameter Summary 1562 11. Appendix B - Tracing and Troubleshooting support 1564 There are several packet types used to gather DVMRP specific 1565 information. They are generally used for diagnosing problems or 1566 gathering topology information. The first two messages are now 1567 obsoleted and should not be used. The remaining two messages provide 1568 a request/response mechanism to determine the versions and 1569 capabilities of a particular DVMRP router. 1571 Code Packet Type Description 1572 ----------------------------------------------------------- 1573 3 DVMRP Ask Neighbors Obsolete 1574 4 DVMRP Neighbors Obsolete 1575 5 DVMRP Ask Neighbors 2 Request Neighbor List 1576 6 DVMRP Neighbors 2 Respond with Neighbor List 1577 ----------------------------------------------------------- 1579 Table 3 - Debugging Packet Types 1581 11.1. DVMRP Ask Neighbors2 1583 The Ask Neighbors2 packet is a unicast request packet directed at a 1584 DVMRP router. The destination should respond with a unicast 1585 Neighbors2 message back to the sender of the Ask Neighbors2 message. 1587 0 8 16 31 1588 +---------+---------+--------------------+ 1589 | Type | Code | Checksum | 1590 |(0x13) | (0x5) | | 1591 +---------+---------+----------+---------+ 1592 | Reserved | Minor | Major | 1593 | | Version |Version | 1594 +-------------------+----------+---------+ 1596 Figure 8 - Ask Neighbors 2 Packet Format 1598 11.2. DVMRP Neighbors2 1600 The format of a Neighbors2 response packet is shown below. This is 1601 sent as a unicast message back to the sender of an Ask Neighbors2 1602 message. There is a common header at the top followed by the routers 1603 capabilities. One or more sections follow that contain an entry for 1604 each logical interface. The interface parameters are listed along 1605 with a variable list of neighbors learned on each interface. 1607 If the interface is down or disabled, list a single neighbor with an 1608 address of 0.0.0.0 for physical interfaces or the remote tunnel 1609 endpoint address for tunnel pseudo-interfaces. 1611 0 8 16 31 1612 +-----------+--------------+--------------------------+ 1613 | Type | Code | Checksum | 1614 | (0x13) | (0x6) | | 1615 +-----------+--------------+------------+-------------+ 1616 | Reserved | Capabilities | Minor | Major | 1617 | | | Version | Version | 1618 +-----------+--------------+------------+-------------+ 1619 | | 1620 | Local Addr 1 | 1621 +-----------+--------------+------------+-------------+ 1622 | | | | | 1623 | Metric 1 | Threshold 1 | Flags 1 | Nbr Count 1 | 1624 +-----------+--------------+------------+-------------+ 1625 | | 1626 | Nbr 1 | 1627 +-----------------------------------------------------+ 1628 | | 1629 | ... | 1630 +-----------------------------------------------------+ 1631 | | 1632 | Nbr m | 1633 +-----------------------------------------------------+ 1634 | | 1635 | Local Addr N | 1636 +-----------+--------------+------------+-------------+ 1637 | | | | | 1638 | Metric N | Threshold N | Flags N | Nbr Count N | 1639 +-----------+--------------+------------+-------------+ 1640 | | 1641 | Nbr 1 | 1642 +-----------------------------------------------------+ 1643 | | 1644 | ... | 1645 +-----------------------------------------------------+ 1646 | | 1647 | Nbr k | 1648 +-----------------------------------------------------+ 1650 Figure 9 - Neighbors 2 Packet Format 1652 The capabilities of the local router are defined as follows: 1654 Bit Flag Description 1655 --------------------------------------------------- 1657 0 Leaf This is a leaf router 1659 1 Prune This router understands pruning 1661 2 GenID This router sends Generation Id's 1663 3 Mtrace This router handles Mtrace requests 1665 4 Snmp This router supports the DVMRP MIB 1666 --------------------------------------------------- 1668 Table 4 - DVMRP Router Capabilities 1670 The flags associated with a particular interface are: 1672 Bit Flag Description 1673 ---------------------------------------------------------- 1675 0 Tunnel Neighbor reached via tunnel 1677 1 Source Route Tunnel uses IP source routing 1679 2 Reserved No longer used 1681 3 Reserved No longer used 1683 4 Down Operational status down 1685 5 Disabled Administrative status down 1687 6 Querier Querier for interface 1689 7 Leaf No downstream neighbors on interface 1690 ---------------------------------------------------------- 1692 Table 5 - DVMRP Interface flags 1694 12. Appendix C - Version Compatibility 1696 There have been two previous major versions of DVMRP with 1697 implementations still in circulation. If the receipt of a Probe 1698 message reveals a major version of 1 or 2, then it can be assumed 1699 that this neighbor does not support pruning or the use of the 1700 Generation ID in the Probe message. However, since these older 1701 implementations are known to safely ignore the Generation ID and 1702 neighbor information in the Probe packet, it is not necessary to send 1703 specially formatted Probe packets to these neighbors. 1705 There were three minor versions (0, 1, and 2) of major version 3 that 1706 did support pruning but did not support the Generation ID or 1707 capability flags. These special cases will have to be accounted for. 1709 Any other minor versions of major version 3 closely compare to this 1710 specification. 1712 In addition, cisco Systems is known to use their software major and 1713 minor release number as the DVMRP major and minor version number. 1714 These will typically be 10 or 11 for the major version number. 1715 Pruning was introduced in Version 11. 1717 Implementations prior to this specification may not wait to send 1718 route reports until probe messages have been received with the 1719 routers address listed. Reports SHOULD be sent to these neighbors 1720 without first requiring a received probe with the routers address in 1721 it as well as reports from these neighbors SHOULD be accepted. 1722 Although, this allows one-way neighbor relationships to occur, it 1723 does maintain backward compatibility. 1725 It may be necessary to form neighbor relationships based solely on 1726 Route Report messages. Neighbor timeout values may need to be 1727 configured to a value greater than the Route Report Interval for 1728 these neighbors. 1730 Implementations that do not monitor Generation ID changes can create 1731 more noticeable black holes when using long prune lifetimes such as 1732 ~2 hours. This happens when a long prune is sent upstream and then 1733 the router that sent the long prune restarts. If the upstream router 1734 ignores the new Generation ID, the prune received by the upstream 1735 router will not be flushed and the downstream router will have no 1736 knowledge of the upstream prune. For this reason, prunes sent 1737 upstream to routers that are known to ignore Generation ID changes 1738 should have short lifetimes. 1740 If the router must run IGMP version 1 on an interface for backwards 1741 compatibility, DVMRP must elect the DVMRP router with the highest IP 1742 address as the IGMP querier. 1744 Some implementations of tools that send DVMRP Ask Neighbors2 requests 1745 and receive Neighbors2 response messages require a neighbor address 1746 of 0.0.0.0 when no neighbors are listed in the response packet. 1747 (Mrinfo) 1749 When DVMRP protocol packets are sent to tunnel endpoints, some 1750 implementations do not accept packets addressed to the All-DVMRP- 1751 Routers address and then encapsulated with the tunnel endpoint 1752 address. Mrouted versions 3.9beta2 and earlier are known to have 1753 this problem. 1755 13. Intellectual Property Rights Notice 1757 The IETF takes no position regarding the validity or scope of any 1758 intellectual property or other rights that might be claimed to 1759 pertain to the implementation or use of the technology described in 1760 this document or the extent to which any license under such rights 1761 might or might not be available; neither does it represent that it 1762 has made any effort to identify any such rights. Information on the 1763 IETF's procedures with respect to rights in standards-track and 1764 standards-related documentation can be found in BCP-11. Copies of 1765 claims of rights made available for publication and any assurances of 1766 licenses to be made available, or the result of an attempt made to 1767 obtain a general license or permission for the use of such 1768 proprietary rights by implementors or users of this specification can 1769 be obtained from the IETF Secretariat. 1771 The IETF invites any interested party to bring to its attention any 1772 copyrights, patents or patent applications, or other proprietary 1773 rights which may cover technology that may be required to practice 1774 this standard. Please address the information to the IETF Executive 1775 Director. 1777 14. Full Copyright Statement 1779 Copyright (C) The Internet Society (date). All Rights Reserved. 1781 This document and translations of it may be copied and furnished to 1782 others, and derivative works that comment on or otherwise explain it 1783 or assist in its implmentation may be prepared, copied, published and 1784 distributed, in whole or in part, without restriction of any kind, 1785 provided that the above copyright notice and this paragraph are 1786 included on all such copies and derivative works. However, this 1787 document itself may not be modified in any way, such as by removing 1788 the copyright notice or references to the Internet Society or other 1789 Internet organizations, except as needed for the purpose of 1790 developing Internet standards in which case the procedures for 1791 copyrights defined in the Internet Standards process must be 1792 followed, or as required to translate it into languages other than 1793 English. 1795 The limited permissions granted above are perpetual and will not be 1796 revoked by the Internet Society or its successors or assigns. 1798 This document and the information contained herein is provided on an 1799 "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING 1800 TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING 1801 BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION 1802 HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF 1803 MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 1805 Table of Contents 1807 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 2 1808 1.1. Requirements Terminology . . . . . . . . . . . . . . . . . . 2 1809 1.2. Reverse Path Multicasting . . . . . . . . . . . . . . . . . 2 1810 1.3. Tunnel Encapsulation . . . . . . . . . . . . . . . . . . . . 2 1811 1.4. Document Overview . . . . . . . . . . . . . . . . . . . . . 3 1812 2. Protocol Overview . . . . . . . . . . . . . . . . . . . . . . 3 1813 2.1. Neighbor Discovery . . . . . . . . . . . . . . . . . . . . . 4 1814 2.2. Source Location . . . . . . . . . . . . . . . . . . . . . . 4 1815 2.3. Dependent Downstream Routers . . . . . . . . . . . . . . . . 5 1816 2.4. Designated Forwarder . . . . . . . . . . . . . . . . . . . . 6 1817 2.5. Building Multicast Trees . . . . . . . . . . . . . . . . . . 6 1818 2.6. Pruning Multicast Trees . . . . . . . . . . . . . . . . . . 7 1819 2.7. Grafting Multicast Trees . . . . . . . . . . . . . . . . . . 8 1820 3. Detailed Protocol Operation . . . . . . . . . . . . . . . . . 8 1821 3.1. Protocol Header . . . . . . . . . . . . . . . . . . . . . . 8 1822 3.2. Probe Messages . . . . . . . . . . . . . . . . . . . . . . . 10 1823 3.3. Multicast Forwarding . . . . . . . . . . . . . . . . . . . . 15 1824 3.4. Route Exchange . . . . . . . . . . . . . . . . . . . . . . . 16 1825 3.5. Pruning . . . . . . . . . . . . . . . . . . . . . . . . . . 25 1826 3.6. Grafting . . . . . . . . . . . . . . . . . . . . . . . . . . 30 1827 3.7. Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . 34 1828 4. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 35 1829 5. Network Management Considerations . . . . . . . . . . . . . . 35 1830 6. Security Considerations . . . . . . . . . . . . . . . . . . . 35 1831 7. References . . . . . . . . . . . . . . . . . . . . . . . . . . 35 1832 8. Author's Address . . . . . . . . . . . . . . . . . . . . . . . 37 1833 9. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 37 1834 10. Appendix A - Constants & Configurable Parameters . . . . . . 38 1835 11. Appendix B - Tracing and Troubleshooting support . . . . . . 39 1836 11.1. DVMRP Ask Neighbors2 . . . . . . . . . . . . . . . . . . . 39 1837 11.2. DVMRP Neighbors2 . . . . . . . . . . . . . . . . . . . . . 40 1838 12. Appendix C - Version Compatibility . . . . . . . . . . . . . 44 1839 13. Intellectual Property Rights Notice . . . . . . . . . . . . . 46 1840 14. Full Copyright Statement . . . . . . . . . . . . . . . . . . 46