User:Howard C. Berkowitz/RIP0

From Citizendium
Jump to: navigation, search

Network Working Group G. Malkin Request for Comments: 2453 Bay Networks Obsoletes: 1723, 1388 November 1998 STD: 56 Category: Standards Track

                            RIP Version 2

Status of this Memo

  This document specifies an Internet standards track protocol for the
  Internet community, and requests discussion and suggestions for
  improvements.  Please refer to the current edition of the "Internet
  Official Protocol Standards" (STD 1) for the standardization state
  and status of this protocol.  Distribution of this memo is unlimited.

Copyright Notice

  Copyright (C) The Internet Society (1998).  All Rights Reserved.


  This document specifies an extension of the Routing Information
  Protocol (RIP), as defined in [1], to expand the amount of useful
  information carried in RIP messages and to add a measure of security.
  A companion document will define the SNMP MIB objects for RIP-2 [2].
  An additional document will define cryptographic security
  improvements for RIP-2 [3].


  I would like to thank the IETF RIP Working Group for their help in
  improving the RIP-2 protocol. Much of the text for the background
  discussions about distance vector protocols and some of the
  descriptions of the operation of RIP were taken from "Routing
  Information Protocol" by C. Hedrick [1]. Some of the final editing on
  the document was done by Scott Bradner.

Malkin Standards Track [Page 1] � RFC 2453 RIP Version 2 November 1998

Table of Contents

  1.  Justification  . . . . . . . . . . . . . . . . . . . . . . . .  3
  2.  Current RIP  . . . . . . . . . . . . . . . . . . . . . . . . .  3
  3.  Basic Protocol . . . . . . . . . . . . . . . . . . . . . . . .  3
  3.1   Introduction   . . . . . . . . . . . . . . . . . . . . . . .  3
  3.2   Limitations of the Protocol  . . . . . . . . . . . . . . . .  5
  3.3.  Organization of this document  . . . . . . . . . . . . . . .  6
  3.4   Distance Vector Algorithms . . . . . . . . . . . . . . . . .  6
  3.4.1    Dealing with changes in topology  . . . . . . . . . . . . 12
  3.4.2    Preventing instability  . . . . . . . . . . . . . . . . . 13
  3.4.3    Split horizon . . . . . . . . . . . . . . . . . . . . . . 15
  3.4.4    Triggered updates . . . . . . . . . . . . . . . . . . . . 17
  3.5   Protocol Specification   . . . . . . . . . . . . . . . . . . 18
  3.6   Message Format . . . . . . . . . . . . . . . . . . . . . . . 20
  3.7   Addressing Considerations  . . . . . . . . . . . . . . . . . 22
  3.8   Timers . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
  3.9   Input Processing . . . . . . . . . . . . . . . . . . . . . . 25
  3.9.1    Request Messages  . . . . . . . . . . . . . . . . . . . . 25
  3.9.2    Response Messages . . . . . . . . . . . . . . . . . . . . 26
  3.10  Output Processing  . . . . . . . . . . . . . . . . . . . . . 28
  3.10.1   Triggered Updates . . . . . . . . . . . . . . . . . . . . 29
  3.10.2   Generating Response Messages. . . . . . . . . . . . . . . 30
  4.  Protocol Extensions  . . . . . . . . . . . . . . . . . . . . . 31
  4.1   Authentication . . . . . . . . . . . . . . . . . . . . . . . 31
  4.2   Route Tag  . . . . . . . . . . . . . . . . . . . . . . . . . 32
  4.3   Subnet Mask  . . . . . . . . . . . . . . . . . . . . . . . . 32
  4.4   Next Hop . . . . . . . . . . . . . . . . . . . . . . . . . . 33
  4.5   Multicasting . . . . . . . . . . . . . . . . . . . . . . . . 33
  4.6   Queries  . . . . . . . . . . . . . . . . . . . . . . . . . . 33
  5.  Compatibility  . . . . . . . . . . . . . . . . . . . . . . . . 34
  5.1   Compatibility Switch . . . . . . . . . . . . . . . . . . . . 34
  5.2   Authentication . . . . . . . . . . . . . . . . . . . . . . . 34
  5.3   Larger Infinity  . . . . . . . . . . . . . . . . . . . . . . 35
  5.4   Addressless Links  . . . . . . . . . . . . . . . . . . . . . 35
  6.  Interaction between version 1 and version 2  . . . . . . . . . 35
  7.  Security Considerations  . . . . . . . . . . . . . . . . . . . 36
  Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
  References . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
  Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 38
  Full Copyright Statement . . . . . . . . . . . . . . . . . . . . . 39

Malkin Standards Track [Page 2] � RFC 2453 RIP Version 2 November 1998

1. Justification

  With the advent of OSPF and IS-IS, there are those who believe that
  RIP is obsolete.  While it is true that the newer IGP routing
  protocols are far superior to RIP, RIP does have some advantages.
  Primarily, in a small network, RIP has very little overhead in terms
  of bandwidth used and configuration and management time.  RIP is also
  very easy to implement, especially in relation to the newer IGPs.
  Additionally, there are many, many more RIP implementations in the
  field than OSPF and IS-IS combined.  It is likely to remain that way
  for some years yet.
  Given that RIP will be useful in many environments for some period of
  time, it is reasonable to increase RIP's usefulness.  This is
  especially true since the gain is far greater than the expense of the

2. Current RIP

  The current RIP-1 message contains the minimal amount of information
  necessary for routers to route messages through a network.  It also
  contains a large amount of unused space, owing to its origins.
  The current RIP-1 protocol does not consider autonomous systems and
  IGP/EGP interactions, subnetting [11], and authentication since
  implementations of these postdate RIP-1.  The lack of subnet masks is
  a particularly serious problem for routers since they need a subnet
  mask to know how to determine a route.  If a RIP-1 route is a network
  route (all non-network bits 0), the subnet mask equals the network
  mask.  However, if some of the non-network bits are set, the router
  cannot determine the subnet mask.  Worse still, the router cannot
  determine if the RIP-1 route is a subnet route or a host route.
  Currently, some routers simply choose the subnet mask of the
  interface over which the route was learned and determine the route
  type from that.

3. Basic Protocol

3.1 Introduction

  RIP is a routing protocol based on the Bellman-Ford (or distance
  vector) algorithm.  This algorithm has been used for routing
  computations in computer networks since the early days of the
  ARPANET.  The particular packet formats and protocol described here
  are based on the program "routed," which is included with the
  Berkeley distribution of Unix.

Malkin Standards Track [Page 3] � RFC 2453 RIP Version 2 November 1998

  In an international network, such as the Internet, it is very
  unlikely that a single routing protocol will used for the entire
  network.  Rather, the network will be organized as a collection of
  Autonomous Systems (AS), each of which will, in general, be
  administered by a single entity.  Each AS will have its own routing
  technology, which may differ among AS's.  The routing protocol used
  within an AS is referred to as an Interior Gateway Protocol (IGP).  A
  separate protocol, called an Exterior Gateway Protocol (EGP), is used
  to transfer routing information among the AS's.  RIP was designed to
  work as an IGP in moderate-size AS's.  It is not intended for use in
  more complex environments.  For information on the context into which
  RIP-1 is expected to fit, see Braden and Postel [6].
  RIP uses one of a class of routing algorithms known as Distance
  Vector algorithms.  The earliest description of this class of
  algorithms known to the author is in Ford and Fulkerson [8].  Because
  of this, they are sometimes known as Ford-Fulkerson algorithms.  The
  term Bellman-Ford is also used, and derives from the fact that the
  formulation is based on Bellman's equation [4].  The presentation in
  this document is closely based on [5].  This document contains a
  protocol specification.  For an introduction to the mathematics of
  routing algorithms, see [1].  The basic algorithms used by this
  protocol were used in computer routing as early as 1969 in the
  ARPANET.  However, the specific ancestry of this protocol is within
  the Xerox network protocols.  The PUP protocols [7] used the Gateway
  Information Protocol to exchange routing information.  A somewhat
  updated version of this protocol was adopted for the Xerox Network
  Systems (XNS) architecture, with the name Routing Information
  Protocol [9].  Berkeley's routed is largely the same as the Routing
  Information Protocol, with XNS addresses replaced by a more general
  address format capable of handling IPv4 and other types of address,
  and with routing updates limited to one every 30 seconds.  Because of
  this similarity, the term Routing Information Protocol (or just RIP)
  is used to refer to both the XNS protocol and the protocol used by
  RIP is intended for use within the IP-based Internet.  The Internet
  is organized into a number of networks connected by special purpose
  gateways known as routers.  The networks may be either point-to-point
  links or more complex networks such as Ethernet or token ring.  Hosts
  and routers are presented with IP datagrams addressed to some host.
  Routing is the method by which the host or router decides where to
  send the datagram.  It may be able to send the datagram directly to
  the destination, if that destination is on one of the networks that
  are directly connected to the host or router.  However, the
  interesting case is when the destination is not directly reachable.

Malkin Standards Track [Page 4] � RFC 2453 RIP Version 2 November 1998

  In this case, the host or router attempts to send the datagram to a
  router that is nearer the destination.  The goal of a routing
  protocol is very simple: It is to supply the information that is
  needed to do routing.

3.2 Limitations of the Protocol

  This protocol does not solve every possible routing problem.  As
  mentioned above, it is primary intended for use as an IGP in networks
  of moderate size.  In addition, the following specific limitations
  are be mentioned:
  - The protocol is limited to networks whose longest path (the
    network's diameter) is 15 hops.  The designers believe that the
    basic protocol design is inappropriate for larger networks.  Note
    that this statement of the limit assumes that a cost of 1 is used
    for each network.  This is the way RIP is normally configured.  If
    the system administrator chooses to use larger costs, the upper
    bound of 15 can easily become a problem.
  - The protocol depends upon "counting to infinity" to resolve certain
    unusual situations. (This will be explained in the next section.)
    If the system of networks has several hundred networks, and a
    routing loop was formed involving all of them, the resolution of
    the loop would require either much time (if the frequency of
    routing updates were limited) or bandwidth (if updates were sent
    whenever changes were detected).  Such a loop would consume a large
    amount of network bandwidth before the loop was corrected.  We
    believe that in realistic cases, this will not be a problem except
    on slow lines.  Even then, the problem will be fairly unusual,
    since various precautions are taken that should prevent these
    problems in most cases.
  - This protocol uses fixed "metrics" to compare alternative routes.
    It is not appropriate for situations where routes need to be chosen
    based on real-time parameters such a measured delay, reliability,
    or load.  The obvious extensions to allow metrics of this type are
    likely to introduce instabilities of a sort that the protocol is
    not designed to handle.

Malkin Standards Track [Page 5] � RFC 2453 RIP Version 2 November 1998

3.3. Organization of this document

  The main body of this document is organized into two parts, which
  occupy the next two sections:
       A conceptual development and justification of distance vector
       algorithms in general.
       The actual protocol description.
  Each of these two sections can largely stand on its own.  Section 3.4
  attempts to give an informal presentation of the mathematical
  underpinnings of the algorithm.  Note that the presentation follows a
  "spiral" method.  An initial, fairly simple algorithm is described.
  Then refinements are added to it in successive sections.  Section 3.5
  is the actual protocol description.  Except where specific references
  are made to section 3.4, it should be possible to implement RIP
  entirely from the specifications given in section 3.5.

3.4 Distance Vector Algorithms

  Routing is the task of finding a path from a sender to a desired
  destination.  In the IP "Internet model" this reduces primarily to a
  matter of finding a series of routers between the source and
  destination networks.  As long as a message or datagram remains on a
  single network or subnet, any forwarding problems are the
  responsibility of technology that is specific to the network.  For
  example, Ethernet and the ARPANET each define a way in which any
  sender can talk to any specified destination within that one network.
  IP routing comes in primarily when messages must go from a sender on
  one network to a destination on a different one.  In that case, the
  message must pass through one or more routers connecting the
  networks.  If the networks are not adjacent, the message may pass
  through several intervening networks, and the routers connecting
  them.  Once the message gets to a router that is on the same network
  as the destination, that network's own technology is used to get to
  the destination.
  Throughout this section, the term "network" is used generically to
  cover a single broadcast network (e.g., an Ethernet), a point to
  point line, or the ARPANET.  The critical point is that a network is
  treated as a single entity by IP.  Either no forwarding decision is
  necessary (as with a point to point line), or that forwarding is done
  in a manner that is transparent to IP, allowing IP to treat the
  entire network as a single fully-connected system (as with an
  Ethernet or the ARPANET).  Note that the term "network" is used in a
  somewhat different way in discussions of IP addressing.  We are using
  the term "network" here to refer to subnets in cases where subnet

Malkin Standards Track [Page 6] � RFC 2453 RIP Version 2 November 1998

  addressing is in use.
  A number of different approaches for finding routes between networks
  are possible.  One useful way of categorizing these approaches is on
  the basis of the type of information the routers need to exchange in
  order to be able to find routes.  Distance vector algorithms are
  based on the exchange of only a small amount of information.  Each
  entity (router or host) that participates in the routing protocol is
  assumed to keep information about all of the destinations within the
  system.  Generally, information about all entities connected to one
  network is summarized by a single entry, which describes the route to
  all destinations on that network.  This summarization is possible
  because as far as IP is concerned, routing within a network is
  invisible.  Each entry in this routing database includes the next
  router to which datagrams destined for the entity should be sent.  In
  addition, it includes a "metric" measuring the total distance to the
  entity.  Distance is a somewhat generalized concept, which may cover
  the time delay in getting messages to the entity, the dollar cost of
  sending messages to it, etc.  Distance vector algorithms get their
  name from the fact that it is possible to compute optimal routes when
  the only information exchanged is the list of these distances.
  Furthermore, information is only exchanged among entities that are
  adjacent, that is, entities that share a common network.
  Although routing is most commonly based on information about
  networks, it is sometimes necessary to keep track of the routes to
  individual hosts.  The RIP protocol makes no formal distinction
  between networks and hosts.  It simply describes exchange of
  information about destinations, which may be either networks or
  hosts.  (Note however, that it is possible for an implementor to
  choose not to support host routes.  See section 3.2.)  In fact, the
  mathematical developments are most conveniently thought of in terms
  of routes from one host or router to another.  When discussing the
  algorithm in abstract terms, it is best to think of a routing entry
  for a network as an abbreviation for routing entries for all of the
  entities connected to that network.  This sort of abbreviation makes
  sense only because we think of networks as having no internal
  structure that is visible at the IP level.  Thus, we will generally
  assign the same distance to every entity in a given network.
  We said above that each entity keeps a routing database with one
  entry for every possible destination in the system.  An actual
  implementation is likely to need to keep the following information
  about each destination:

Malkin Standards Track [Page 7] � RFC 2453 RIP Version 2 November 1998

  - address: in IP implementations of these algorithms, this will be
    the IP address of the host or network.
  - router: the first router along the route to the destination.
  - interface: the physical network which must be used to reach the
    first router.
  - metric: a number, indicating the distance to the destination.
  - timer: the amount of time since the entry was last updated.
  In addition, various flags and other internal information will
  probably be included.  This database is initialized with a
  description of the entities that are directly connected to the
  system.  It is updated according to information received in messages
  from neighboring routers.
  The most important information exchanged by the hosts and routers is
  carried in update messages.  Each entity that participates in the
  routing scheme sends update messages that describe the routing
  database as it currently exists in that entity.  It is possible to
  maintain optimal routes for the entire system by using only
  information obtained from neighboring entities.  The algorithm used
  for that will be described in the next section.
  As we mentioned above, the purpose of routing is to find a way to get
  datagrams to their ultimate destinations.  Distance vector algorithms
  are based on a table in each router listing the best route to every
  destination in the system.  Of course, in order to define which route
  is best, we have to have some way of measuring goodness.  This is
  referred to as the "metric".
  In simple networks, it is common to use a metric that simply counts
  how many routers a message must go through.  In more complex
  networks, a metric is chosen to represent the total amount of delay
  that the message suffers, the cost of sending it, or some other
  quantity which may be minimized.  The main requirement is that it
  must be possible to represent the metric as a sum of "costs" for
  individual hops.
  Formally, if it is possible to get from entity i to entity j directly
  (i.e., without passing through another router between), then a cost,
  d(i,j), is associated with the hop between i and j.  In the normal
  case where all entities on a given network are considered to be the
  same, d(i,j) is the same for all destinations on a given network, and
  represents the cost of using that network.  To get the metric of a
  complete route, one just adds up the costs of the individual hops

Malkin Standards Track [Page 8] � RFC 2453 RIP Version 2 November 1998

  that make up the route.  For the purposes of this memo, we assume
  that the costs are positive integers.
  Let D(i,j) represent the metric of the best route from entity i to
  entity j.  It should be defined for every pair of entities.  d(i,j)
  represents the costs of the individual steps.  Formally, let d(i,j)
  represent the cost of going directly from entity i to entity j.  It
  is infinite if i and j are not immediate neighbors. (Note that d(i,i)
  is infinite.  That is, we don't consider there to be a direct
  connection from a node to itself.)  Since costs are additive, it is
  easy to show that the best metric must be described by
     D(i,i) = 0,                      all i
     D(i,j) = min [d(i,k) + D(k,j)],  otherwise
  and that the best routes start by going from i to those neighbors k
  for which d(i,k) + D(k,j) has the minimum value.  (These things can
  be shown by induction on the number of steps in the routes.)  Note
  that we can limit the second equation to k's that are immediate
  neighbors of i.  For the others, d(i,k) is infinite, so the term
  involving them can never be the minimum.
  It turns out that one can compute the metric by a simple algorithm
  based on this.  Entity i gets its neighbors k to send it their
  estimates of their distances to the destination j.  When i gets the
  estimates from k, it adds d(i,k) to each of the numbers.  This is
  simply the cost of traversing the network between i and k.  Now and
  then i compares the values from all of its neighbors and picks the
  A proof is given in [2] that this algorithm will converge to the
  correct estimates of D(i,j) in finite time in the absence of topology
  changes.  The authors make very few assumptions about the order in
  which the entities send each other their information, or when the min
  is recomputed.  Basically, entities just can't stop sending updates
  or recomputing metrics, and the networks can't delay messages
  forever.  (Crash of a routing entity is a topology change.)  Also,
  their proof does not make any assumptions about the initial estimates
  of D(i,j), except that they must be non-negative.  The fact that
  these fairly weak assumptions are good enough is important.  Because
  we don't have to make assumptions about when updates are sent, it is
  safe to run the algorithm asynchronously.  That is, each entity can
  send updates according to its own clock.  Updates can be dropped by
  the network, as long as they don't all get dropped.  Because we don't
  have to make assumptions about the starting condition, the algorithm
  can handle changes.  When the system changes, the routing algorithm
  starts moving to a new equilibrium, using the old one as its starting
  point.  It is important that the algorithm will converge in finite

Malkin Standards Track [Page 9] � RFC 2453 RIP Version 2 November 1998

  time no matter what the starting point.  Otherwise certain kinds of
  changes might lead to non-convergent behavior.
  The statement of the algorithm given above (and the proof) assumes
  that each entity keeps copies of the estimates that come from each of
  its neighbors, and now and then does a min over all of the neighbors.
  In fact real implementations don't necessarily do that.  They simply
  remember the best metric seen so far, and the identity of the
  neighbor that sent it.  They replace this information whenever they
  see a better (smaller) metric.  This allows them to compute the
  minimum incrementally, without having to store data from all of the
  There is one other difference between the algorithm as described in
  texts and those used in real protocols such as RIP: the description
  above would have each entity include an entry for itself, showing a
  distance of zero.  In fact this is not generally done.  Recall that
  all entities on a network are normally summarized by a single entry
  for the network.  Consider the situation of a host or router G that
  is connected to network A.  C represents the cost of using network A
  (usually a metric of one).  (Recall that we are assuming that the
  internal structure of a network is not visible to IP, and thus the
  cost of going between any two entities on it is the same.)  In
  principle, G should get a message from every other entity H on
  network A, showing a cost of 0 to get from that entity to itself.  G
  would then compute C + 0 as the distance to H.  Rather than having G
  look at all of these identical messages, it simply starts out by
  making an entry for network A in its table, and assigning it a metric
  of C.  This entry for network A should be thought of as summarizing
  the entries for all other entities on network A.  The only entity on
  A that can't be summarized by that common entry is G itself, since
  the cost of going from G to G is 0, not C.  But since we never need
  those 0 entries, we can safely get along with just the single entry
  for network A.  Note one other implication of this strategy: because
  we don't need to use the 0 entries for anything, hosts that do not
  function as routers don't need to send any update messages.  Clearly
  hosts that don't function as routers (i.e., hosts that are connected
  to only one network) can have no useful information to contribute
  other than their own entry D(i,i) = 0.  As they have only the one
  interface, it is easy to see that a route to any other network
  through them will simply go in that interface and then come right
  back out it.  Thus the cost of such a route will be greater than the
  best cost by at least C.  Since we don't need the 0 entries, non-
  routers need not participate in the routing protocol at all.
  Let us summarize what a host or router G does.  For each destination
  in the system, G will keep a current estimate of the metric for that
  destination (i.e., the total cost of getting to it) and the identity

Malkin Standards Track [Page 10] � RFC 2453 RIP Version 2 November 1998

  of the neighboring router on whose data that metric is based.  If the
  destination is on a network that is directly connected to G, then G
  simply uses an entry that shows the cost of using the network, and
  the fact that no router is needed to get to the destination.  It is
  easy to show that once the computation has converged to the correct
  metrics, the neighbor that is recorded by this technique is in fact
  the first router on the path to the destination.  (If there are
  several equally good paths, it is the first router on one of them.)
  This combination of destination, metric, and router is typically
  referred to as a route to the destination with that metric, using
  that router. The method so far only has a way to lower the metric, as the
  existing metric is kept until a smaller one shows up.  It is possible
  that the initial estimate might be too low.  Thus, there must be a
  way to increase the metric.  It turns out to be sufficient to use the
  following rule: suppose the current route to a destination has metric
  D and uses router G.  If a new set of information arrived from some
  source other than G, only update the route if the new metric is
  better than D.  But if a new set of information arrives from G
  itself, always update D to the new value.  It is easy to show that
  with this rule, the incremental update process produces the same
  routes as a calculation that remembers the latest information from
  all the neighbors and does an explicit minimum.  (Note that the
  discussion so far assumes that the network configuration is static.
  It does not allow for the possibility that a system might fail.)
  To summarize, here is the basic distance vector algorithm as it has
  been developed so far.  (Note that this is not a statement of the RIP
  protocol.  There are several refinements still to be added.)  The
  following procedure is carried out by every entity that participates
  in the routing protocol.  This must include all of the routers in the
  system.  Hosts that are not routers may participate as well.
  - Keep a table with an entry for every possible destination in the
    system.  The entry contains the distance D to the destination, and
    the first router G on the route to that network.  Conceptually,
    there should be an entry for the entity itself, with metric 0, but
    this is not actually included.
  - Periodically, send a routing update to every neighbor.  The update
    is a set of messages that contain all of the information from the
    routing table.  It contains an entry for each destination, with the
    distance shown to that destination.
  - When a routing update arrives from a neighbor G', add the cost
    associated with the network that is shared with G'.  (This should
    be the network over which the update arrived.)  Call the resulting

Malkin Standards Track [Page 11] � RFC 2453 RIP Version 2 November 1998

    distance D'.  Compare the resulting distances with the current
    routing table entries.  If the new distance D' for N is smaller
    than the existing value D, adopt the new route.  That is, change
    the table entry for N to have metric D' and router G'.  If G' is
    the router from which the existing route came, i.e., G' = G, then
    use the new metric even if it is larger than the old one.

3.4.1 Dealing with changes in topology

  The discussion above assumes that the topology of the network is
  fixed.  In practice, routers and lines often fail and come back up.
  To handle this possibility, we need to modify the algorithm slightly.
  The theoretical version of the algorithm involved a minimum over all
  immediate neighbors.  If the topology changes, the set of neighbors
  changes.  Therefore, the next time the calculation is done, the
  change will be reflected.  However, as mentioned above, actual
  implementations use an incremental version of the minimization.  Only
  the best route to any given destination is remembered.  If the router
  involved in that route should crash, or the network connection to it
  break, the calculation might never reflect the change.  The algorithm
  as shown so far depends upon a router notifying its neighbors if its
  metrics change.  If the router crashes, then it has no way of
  notifying neighbors of a change.
  In order to handle problems of this kind, distance vector protocols
  must make some provision for timing out routes.  The details depend
  upon the specific protocol.  As an example, in RIP every router that
  participates in routing sends an update message to all its neighbors
  once every 30 seconds.  Suppose the current route for network N uses
  router G.  If we don't hear from G for 180 seconds, we can assume
  that either the router has crashed or the network connecting us to it
  has become unusable.  Thus, we mark the route as invalid.  When we
  hear from another neighbor that has a valid route to N, the valid
  route will replace the invalid one.  Note that we wait for 180
  seconds before timing out a route even though we expect to hear from
  each neighbor every 30 seconds.  Unfortunately, messages are
  occasionally lost by networks.  Thus, it is probably not a good idea
  to invalidate a route based on a single missed message.
  As we will see below, it is useful to have a way to notify neighbors
  that there currently isn't a valid route to some network.  RIP, along
  with several other protocols of this class, does this through a
  normal update message, by marking that network as unreachable.  A
  specific metric value is chosen to indicate an unreachable
  destination; that metric value is larger than the largest valid
  metric that we expect to see.  In the existing implementation of RIP,
  16 is used.  This value is normally referred to as "infinity", since

Malkin Standards Track [Page 12] � RFC 2453 RIP Version 2 November 1998

  it is larger than the largest valid metric.  16 may look like a
  surprisingly small number.  It is chosen to be this small for reasons
  that we will see shortly.  In most implementations, the same
  convention is used internally to flag a route as invalid.

3.4.2 Preventing instability

  The algorithm as presented up to this point will always allow a host
  or router to calculate a correct routing table.  However, that is
  still not quite enough to make it useful in practice.  The proofs
  referred to above only show that the routing tables will converge to
  the correct values in finite time.  They do not guarantee that this
  time will be small enough to be useful, nor do they say what will
  happen to the metrics for networks that become inaccessible.
  It is easy enough to extend the mathematics to handle routes becoming
  inaccessible.  The convention suggested above will do that.  We
  choose a large metric value to represent "infinity".  This value must
  be large enough that no real metric would ever get that large.  For
  the purposes of this example, we will use the value 16.  Suppose a
  network becomes inaccessible.  All of the immediately neighboring
  routers time out and set the metric for that network to 16.  For
  purposes of analysis, we can assume that all the neighboring routers
  have gotten a new piece of hardware that connects them directly to
  the vanished network, with a cost of 16.  Since that is the only
  connection to the vanished network, all the other routers in the
  system will converge to new routes that go through one of those
  routers.  It is easy to see that once convergence has happened, all
  the routers will have metrics of at least 16 for the vanished
  network.  Routers one hop away from the original neighbors would end
  up with metrics of at least 17; routers two hops away would end up
  with at least 18, etc.  As these metrics are larger than the maximum
  metric value, they are all set to 16.  It is obvious that the system
  will now converge to a metric of 16 for the vanished network at all
  Unfortunately, the question of how long convergence will take is not
  amenable to quite so simple an answer.  Before going any further, it
  will be useful to look at an example (taken from [2]).  Note that
  what we are about to show will not happen with a correct
  implementation of RIP.  We are trying to show why certain features
  are needed.  In the following example the letters correspond to
  routers, and the lines to networks.

Malkin Standards Track [Page 13] � RFC 2453 RIP Version 2 November 1998

     \   / \
      \ /  |
       C  /    all networks have cost 1, except
       | /     for the direct link from C to D, which
       |/      has cost 10
       |<=== target network
  Each router will have a table showing a route to each network.
  However, for purposes of this illustration, we show only the routes
  from each router to the network marked at the bottom of the diagram.
          D:  directly connected, metric 1
          B:  route via D, metric 2
          C:  route via B, metric 3
          A:  route via B, metric 3
  Now suppose that the link from B to D fails.  The routes should now
  adjust to use the link from C to D.  Unfortunately, it will take a
  while for this to this to happen.  The routing changes start when B
  notices that the route to D is no longer usable.  For simplicity, the
  chart below assumes that all routers send updates at the same time.
  The chart shows the metric for the target network, as it appears in
  the routing table at each router.
      time ------>
      D: dir, 1   dir, 1   dir, 1   dir, 1  ...  dir, 1   dir, 1
      B: unreach  C,   4   C,   5   C,   6       C,  11   C,  12
      C: B,   3   A,   4   A,   5   A,   6       A,  11   D,  11
      A: B,   3   C,   4   C,   5   C,   6       C,  11   C,  12
      dir = directly connected
      unreach = unreachable
  Here's the problem:  B is able to get rid of its failed route using a
  timeout mechanism, but vestiges of that route persist in the system
  for a long time.  Initially, A and C still think they can get to D
  via B.  So, they keep sending updates listing metrics of 3.  In the
  next iteration, B will then claim that it can get to D via either A
  or C.  Of course, it can't.  The routes being claimed by A and C are
  now gone, but they have no way of knowing that yet.  And even when
  they discover that their routes via B have gone away, they each think
  there is a route available via the other.  Eventually the system
  converges, as all the mathematics claims it must.  But it can take
  some time to do so.  The worst case is when a network becomes

Malkin Standards Track [Page 14] � RFC 2453 RIP Version 2 November 1998

  completely inaccessible from some part of the system.  In that case,
  the metrics may increase slowly in a pattern like the one above until
  they finally reach infinity.  For this reason, the problem is called
  "counting to infinity".
  You should now see why "infinity" is chosen to be as small as
  possible.  If a network becomes completely inaccessible, we want
  counting to infinity to be stopped as soon as possible.  Infinity
  must be large enough that no real route is that big.  But it
  shouldn't be any bigger than required.  Thus the choice of infinity
  is a tradeoff between network size and speed of convergence in case
  counting to infinity happens.  The designers of RIP believed that the
  protocol was unlikely to be practical for networks with a diameter
  larger than 15.
  There are several things that can be done to prevent problems like
  this.  The ones used by RIP are called "split horizon with poisoned
  reverse", and "triggered updates".

3.4.3 Split horizon

  Note that some of the problem above is caused by the fact that A and
  C are engaged in a pattern of mutual deception.  Each claims to be
  able to get to D via the other.  This can be prevented by being a bit
  more careful about where information is sent.  In particular, it is
  never useful to claim reachability for a destination network to the
  neighbor(s) from which the route was learned.  "Split horizon" is a
  scheme for avoiding problems caused by including routes in updates
  sent to the router from which they were learned.  The "simple split
  horizon" scheme omits routes learned from one neighbor in updates
  sent to that neighbor.  "Split horizon with poisoned reverse"
  includes such routes in updates, but sets their metrics to infinity.
  If A thinks it can get to D via C, its messages to C should indicate
  that D is unreachable.  If the route through C is real, then C either
  has a direct connection to D, or a connection through some other
  router.  C's route can't possibly go back to A, since that forms a
  loop.  By telling C that D is unreachable, A simply guards against
  the possibility that C might get confused and believe that there is a
  route through A.  This is obvious for a point to point line.  But
  consider the possibility that A and C are connected by a broadcast
  network such as an Ethernet, and there are other routers on that
  network.  If A has a route through C, it should indicate that D is
  unreachable when talking to any other router on that network.  The
  other routers on the network can get to C themselves.  They would
  never need to get to C via A.  If A's best route is really through C,
  no other router on that network needs to know that A can reach D.
  This is fortunate, because it means that the same update message that

Malkin Standards Track [Page 15] � RFC 2453 RIP Version 2 November 1998

  is used for C can be used for all other routers on the same network.
  Thus, update messages can be sent by broadcast.
  In general, split horizon with poisoned reverse is safer than simple
  split horizon.  If two routers have routes pointing at each other,
  advertising reverse routes with a metric of 16 will break the loop
  immediately.  If the reverse routes are simply not advertised, the
  erroneous routes will have to be eliminated by waiting for a timeout.
  However, poisoned reverse does have a disadvantage: it increases the
  size of the routing messages.  Consider the case of a campus backbone
  connecting a number of different buildings.  In each building, there
  is a router connecting the backbone to a local network.  Consider
  what routing updates those routers should broadcast on the backbone
  network.  All that the rest of the network really needs to know about
  each router is what local networks it is connected to.  Using simple
  split horizon, only those routes would appear in update messages sent
  by the router to the backbone network.  If split horizon with
  poisoned reverse is used, the router must mention all routes that it
  learns from the backbone, with metrics of 16.  If the system is
  large, this can result in a large update message, almost all of whose
  entries indicate unreachable networks.
  In a static sense, advertising reverse routes with a metric of 16
  provides no additional information.  If there are many routers on one
  broadcast network, these extra entries can use significant bandwidth.
  The reason they are there is to improve dynamic behavior.  When
  topology changes, mentioning routes that should not go through the
  router as well as those that should can speed up convergence.
  However, in some situations, network managers may prefer to accept
  somewhat slower convergence in order to minimize routing overhead.
  Thus implementors may at their option implement simple split horizon
  rather than split horizon with poisoned reverse, or they may provide
  a configuration option that allows the network manager to choose
  which behavior to use.  It is also permissible to implement hybrid
  schemes that advertise some reverse routes with a metric of 16 and
  omit others.  An example of such a scheme would be to use a metric of
  16 for reverse routes for a certain period of time after routing
  changes involving them, and thereafter omitting them from updates.
  The router requirements RFC [11] specifies that all implementation of
  RIP must use split horizon and should also use split horizon with
  poisoned reverse, although there may be a knob to disable poisoned

Malkin Standards Track [Page 16] � RFC 2453 RIP Version 2 November 1998

3.4.4 Triggered updates

  Split horizon with poisoned reverse will prevent any routing loops
  that involve only two routers.  However, it is still possible to end
  up with patterns in which three routers are engaged in mutual
  deception.  For example, A may believe it has a route through B, B
  through C, and C through A.  Split horizon cannot stop such a loop.
  This loop will only be resolved when the metric reaches infinity and
  the network involved is then declared unreachable.  Triggered updates
  are an attempt to speed up this convergence.  To get triggered
  updates, we simply add a rule that whenever a router changes the
  metric for a route, it is required to send update messages almost
  immediately, even if it is not yet time for one of the regular update
  message.  (The timing details will differ from protocol to protocol.
  Some distance vector protocols, including RIP, specify a small time
  delay, in order to avoid having triggered updates generate excessive
  network traffic.)  Note how this combines with the rules for
  computing new metrics.  Suppose a router's route to destination N
  goes through router G.  If an update arrives from G itself, the
  receiving router is required to believe the new information, whether
  the new metric is higher or lower than the old one.  If the result is
  a change in metric, then the receiving router will send triggered
  updates to all the hosts and routers directly connected to it.  They
  in turn may each send updates to their neighbors.  The result is a
  cascade of triggered updates.  It is easy to show which routers and
  hosts are involved in the cascade.  Suppose a router G times out a
  route to destination N.  G will send triggered updates to all of its
  neighbors.  However, the only neighbors who will believe the new
  information are those whose routes for N go through G.  The other
  routers and hosts will see this as information about a new route that
  is worse than the one they are already using, and ignore it.  The
  neighbors whose routes go through G will update their metrics and
  send triggered updates to all of their neighbors.  Again, only those
  neighbors whose routes go through them will pay attention.  Thus, the
  triggered updates will propagate backwards along all paths leading to
  router G, updating the metrics to infinity.  This propagation will
  stop as soon as it reaches a portion of the network whose route to
  destination N takes some other path.
  If the system could be made to sit still while the cascade of
  triggered updates happens, it would be possible to prove that
  counting to infinity will never happen.  Bad routes would always be
  removed immediately, and so no routing loops could form.
  Unfortunately, things are not so nice.  While the triggered updates
  are being sent, regular updates may be happening at the same time.
  Routers that haven't received the triggered update yet will still be
  sending out information based on the route that no longer exists.  It

Malkin Standards Track [Page 17] � RFC 2453 RIP Version 2 November 1998

  is possible that after the triggered update has gone through a
  router, it might receive a normal update from one of these routers
  that hasn't yet gotten the word.  This could reestablish an orphaned
  remnant of the faulty route.  If triggered updates happen quickly
  enough, this is very unlikely.  However, counting to infinity is
  still possible.
  The router requirements RFC [11] specifies that all implementation of
  RIP must implement triggered update for deleted routes and may
  implement triggered updates for new routes or change of routes.  RIP
  implementations must also limit the rate which of triggered updates
  may be trandmitted. (see section 3.10.1)

3.5 Protocol Specification

  RIP is intended to allow routers to exchange information for
  computing routes through an IPv4-based network.  Any router that uses
  RIP is assumed to have interfaces to one or more networks, otherwise
  it isn't really a router.  These are referred to as its directly-
  connected networks.  The protocol relies on access to certain
  information about each of these networks, the most important of which
  is its metric.  The RIP metric of a network is an integer between 1
  and 15, inclusive.  It is set in some manner not specified in this
  protocol; however, given the maximum path limit of 15, a value of 1
  is usually used.  Implementations should allow the system
  administrator to set the metric of each network.  In addition to the
  metric, each network will have an IPv4 destination address and subnet
  mask associated with it.  These are to be set by the system
  administrator in a manner not specified in this protocol.
  Any host that uses RIP is assumed to have interfaces to one or more
  networks.  These are referred to as its "directly-connected
  networks".  The protocol relies on access to certain information
  about each of these networks.  The most important is its metric or
  "cost".  The metric of a network is an integer between 1 and 15
  inclusive.  It is set in some manner not specified in this protocol.
  Most existing implementations always use a metric of 1.  New
  implementations should allow the system administrator to set the cost
  of each network.  In addition to the cost, each network will have an
  IPv4 network number and a subnet mask associated with it.  These are
  to be set by the system administrator in a manner not specified in
  this protocol.
  Note that the rules specified in section 3.7 assume that there is a
  single subnet mask applying to each IPv4 network, and that only the
  subnet masks for directly-connected networks are known.  There may be
  systems that use different subnet masks for different subnets within
  a single network.  There may also be instances where it is desirable

Malkin Standards Track [Page 18] � RFC 2453 RIP Version 2 November 1998

  for a system to know the subnets masks of distant networks. Network-
  wide distribution of routing information which contains different
  subnet masks is permitted if all routers in the network are running
  the extensions presented in this document. However, if all routers in
  the network are not running these extensions distribution of routing
  information containing different subnet masks must be limited to
  avoid interoperability problems. See sections 3.7 and 4.3 for the
  rules governing subnet distribution.
  Each router that implements RIP is assumed to have a routing table.
  This table has one entry for every destination that is reachable
  throughout the system operating RIP.  Each entry contains at least
  the following information:
  - The IPv4 address of the destination.
  - A metric, which represents the total cost of getting a datagram
    from the router to that destination.  This metric is the sum of the
    costs associated with the networks that would be traversed to get
    to the destination.
  - The IPv4 address of the next router along the path to the
    destination (i.e., the next hop).  If the destination is on one of
    the directly-connected networks, this item is not needed.
  - A flag to indicate that information about the route has changed
    recently.  This will be referred to as the "route change flag."
  - Various timers associated with the route.  See section 3.6 for more
    details on timers.
  The entries for the directly-connected networks are set up by the
  router using information gathered by means not specified in this
  protocol.  The metric for a directly-connected network is set to the
  cost of that network.  As mentioned, 1 is the usual cost.  In that
  case, the RIP metric reduces to a simple hop-count.  More complex
  metrics may be used when it is desirable to show preference for some
  networks over others (e.g., to indicate of differences in bandwidth
  or reliability).
  To support the extensions detailed in this document, each entry must
  additionally contain a subnet mask. The subnet mask allows the router
  (along with the IPv4 address of the destination) to identify the
  different subnets within a single network as well as the subnets
  masks of distant networks.

Malkin Standards Track [Page 19] � RFC 2453 RIP Version 2 November 1998

  Implementors may also choose to allow the system administrator to
  enter additional routes.  These would most likely be routes to hosts
  or networks outside the scope of the routing system.  They are
  referred to as "static routes."  Entries for destinations other than
  these initial ones are added and updated by the algorithms described
  in the following sections.
  In order for the protocol to provide complete information on routing,
  every router in the AS must participate in the protocol.  In cases
  where multiple IGPs are in use, there must be at least one router
  which can leak routing information between the protocols.

3.6 Message Format

  RIP is a UDP-based protocol.  Each router that uses RIP has a routing
  process that sends and receives datagrams on UDP port number 520, the
  RIP-1/RIP-2 port.  All communications intended for another routers's
  RIP process are sent to the RIP port.  All routing update messages
  are sent from the RIP port.  Unsolicited routing update messages have
  both the source and destination port equal to the RIP port.  Update
  messages sent in response to a request are sent to the port from
  which the request came.  Specific queries may be sent from ports
  other than the RIP port, but they must be directed to the RIP port on
  the target machine.
  The RIP packet format is:
      0                   1                   2                   3
      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
     |  command (1)  |  version (1)  |       must be zero (2)        |
     |                                                               |
     ~                         RIP Entry (20)                        ~
     |                                                               |

Malkin Standards Track [Page 20] � RFC 2453 RIP Version 2 November 1998

  There may be between 1 and 25 (inclusive) RIP entries.  A RIP-1 entry
  has the following format:
      0                   1                   2                   3
      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
     | address family identifier (2) |      must be zero (2)         |
     |                        IPv4 address (4)                       |
     |                        must be zero (4)                       |
     |                        must be zero (4)                       |
     |                           metric (4)                          |
  Field sizes are given in octets.  Unless otherwise specified, fields
  contain binary integers, in network byte order, with the most-
  significant octet first (big-endian).  Each tick mark represents one
  Every message contains a RIP header which consists of a command and a
  version number.  This section of the document describes version 1 of
  the protocol; section 4 describes the version 2 extensions.  The
  command field is used to specify the purpose of this message.  The
  commands implemented in version 1 and 2 are:
  1 - request    A request for the responding system to send all or
                 part of its routing table.
  2 - response   A message containing all or part of the sender's
                 routing table.  This message may be sent in response
                 to a request, or it may be an unsolicited routing
                 update generated by the sender.
  For each of these message types, in version 1, the remainder of the
  datagram contains a list of Route Entries (RTEs).  Each RTE in this
  list contains an Address Family Identifier (AFI), destination IPv4
  address, and the cost to reach that destination (metric).
  The AFI is the type of address.  For RIP-1, only AF_INET (2) is
  generally supported.
  The metric field contains a value between 1 and 15 (inclusive) which
  specifies the current metric for the destination; or the value 16
  (infinity), which indicates that the destination is not reachable.

Malkin Standards Track [Page 21] � RFC 2453 RIP Version 2 November 1998

3.7 Addressing Considerations

  Distance vector routing can be used to describe routes to individual
  hosts or to networks.  The RIP protocol allows either of these
  possibilities.  The destinations appearing in request and response
  messages can be networks, hosts, or a special code used to indicate a
  default address.  In general, the kinds of routes actually used will
  depend upon the routing strategy used for the particular network.
  Many networks are set up so that routing information for individual
  hosts is not needed.  If every node on a given network or subnet is
  accessible through the same routers, then there is no reason to
  mention individual hosts in the routing tables.  However, networks
  that include point-to-point lines sometimes require routers to keep
  track of routes to certain nodes.  Whether this feature is required
  depends upon the addressing and routing approach used in the system.
  Thus, some implementations may choose not to support host routes.  If
  host routes are not supported, they are to be dropped when they are
  received in response messages (see section 3.7.2).
  The RIP-1 packet format does not distinguish among various types of
  address.  Fields that are labeled "address" can contain any of the
  host address subnet number network number zero (default route)
  Entities which use RIP-1 are assumed to use the most specific
  information available when routing a datagram.  That is, when routing
  a datagram, its destination address must first be checked against the
  list of node addresses.  Then it must be checked to see whether it
  matches any known subnet or network number.  Finally, if none of
  these match, the default route is used.
  When a node evaluates information that it receives via RIP-1, its
  interpretation of an address depends upon whether it knows the subnet
  mask that applies to the net.  If so, then it is possible to
  determine the meaning of the address.  For example, consider net
  128.6.  It has a subnet mask of  Thus is a
  network number, is a subnet number, and is a node
  address.  However, if the node does not know the subnet mask,
  evaluation of an address may be ambiguous.  If there is a non-zero
  node part, there is no clear way to determine whether the address
  represents a subnet number or a node address.  As a subnet number
  would be useless without the subnet mask, addresses are assumed to
  represent nodes in this situation.  In order to avoid this sort of
  ambiguity, when using version 1, nodes must not send subnet routes to
  nodes that cannot be expected to know the appropriate subnet mask.
  Normally hosts only know the subnet masks for directly-connected
  networks.  Therefore, unless special provisions have been made,

Malkin Standards Track [Page 22] � RFC 2453 RIP Version 2 November 1998

  routes to a subnet must not be sent outside the network of which the
  subnet is a part.  RIP-2 (see section 4) eliminates the subnet/host
  ambiguity by including the subnet mask in the routing entry.
  This "subnet filtering" is carried out by the routers at the "border"
  of the subnetted network.  These are routers which connect that
  network with some other network.  Within the subnetted network, each
  subnet is treated as an individual network.  Routing entries for each
  subnet are circulated by RIP.  However, border routers send only a
  single entry for the network as a whole to nodes in other networks.
  This means that a border router will send different information to
  different neighbors.  For neighbors connected to the subnetted
  network, it generates a list of all subnets to which it is directly
  connected, using the subnet number.  For neighbors connected to other
  networks, it makes a single entry for the network as a whole, showing
  the metric associated with that network.  This metric would normally
  be the smallest metric for the subnets to which the router is
  Similarly, border routers must not mention host routes for nodes
  within one of the directly-connected networks in messages to other
  networks.  Those routes will be subsumed by the single entry for the
  network as a whole.
  The router requirements RFC [11] specifies that all implementation of
  RIP should support host routes but if they do not then they must
  ignore any received host routes.
  The special address is used to describe a default route.  A
  default route is used when it is not convenient to list every
  possible network in the RIP updates, and when one or more closely-
  connected routers in the system are prepared to handle traffic to the
  networks that are not listed explicitly.  These routers should create
  RIP entries for the address, just as if it were a network to
  which they are connected.  The decision as to how routers create
  entries for is left to the implementor.  Most commonly, the
  system administrator will be provided with a way to specify which
  routers should create entries for; however, other mechanisms
  are possible.  For example, an implementor might decide that any
  router which speaks BGP should be declared to be a default router.
  It may be useful to allow the network administrator to choose the
  metric to be used in these entries.  If there is more than one
  default router, this will make it possible to express a preference
  for one over the other.  The entries for are handled by RIP
  in exactly the same manner as if there were an actual network with
  this address.  System administrators should take care to make sure
  that routes to do not propagate further than is intended.
  Generally, each autonomous system has its own preferred default

Malkin Standards Track [Page 23] � RFC 2453 RIP Version 2 November 1998

  router.  Thus, routes involving should generally not leave
  the boundary of an autonomous system.  The mechanisms for enforcing
  this are not specified in this document.

3.8 Timers

  This section describes all events that are triggered by timers.
  Every 30 seconds, the RIP process is awakened to send an unsolicited
  Response message containing the complete routing table (see section
  3.9 on Split Horizon) to every neighboring router.  When there are
  many routers on a single network, there is a tendency for them to
  synchronize with each other such that they all issue updates at the
  same time.  This can happen whenever the 30 second timer is affected
  by the processing load on the system.  It is undesirable for the
  update messages to become synchronized, since it can lead to
  unnecessary collisions on broadcast networks.  Therefore,
  implementations are required to take one of two precautions:
  - The 30-second updates are triggered by a clock whose rate is not
    affected by system load or the time required to service the
    previous update timer.
  - The 30-second timer is offset by a small random time (+/- 0 to 5
    seconds) each time it is set.  (Implementors may wish to consider
    even larger variation in the light of recent research results [10])
  There are two timers associated with each route, a "timeout" and a
  "garbage-collection" time.  Upon expiration of the timeout, the route
  is no longer valid; however, it is retained in the routing table for
  a short time so that neighbors can be notified that the route has
  been dropped.  Upon expiration of the garbage-collection timer, the
  route is finally removed from the routing table.
  The timeout is initialized when a route is established, and any time
  an update message is received for the route.  If 180 seconds elapse
  from the last time the timeout was initialized, the route is
  considered to have expired, and the deletion process described below
  begins for that route.
  Deletions can occur for one of two reasons: the timeout expires, or
  the metric is set to 16 because of an update received from the
  current router (see section 3.7.2 for a discussion of processing
  updates from other routers).  In either case, the following events

Malkin Standards Track [Page 24] � RFC 2453 RIP Version 2 November 1998

  - The garbage-collection timer is set for 120 seconds.
  - The metric for the route is set to 16 (infinity).  This causes the
    route to be removed from service.
  - The route change flag is set to indicate that this entry has been
  - The output process is signalled to trigger a response.
  Until the garbage-collection timer expires, the route is included in
  all updates sent by this router.  When the garbage-collection timer
  expires, the route is deleted from the routing table.
  Should a new route to this network be established while the garbage-
  collection timer is running, the new route will replace the one that
  is about to be deleted.  In this case the garbage-collection timer
  must be cleared.
  Triggered updates also use a small timer; however, this is best
  described in section 3.9.1.

3.9 Input Processing

  This section will describe the handling of datagrams received on the
  RIP port.  Processing will depend upon the value in the command
  See sections 4.6 and 5.1 for details on handling version numbers.

3.9.1 Request Messages

  A Request is used to ask for a response containing all or part of a
  router's routing table.  Normally, Requests are sent as broadcasts
  (multicasts for RIP-2), from the RIP port, by routers which have just
  come up and are seeking to fill in their routing tables as quickly as
  possible.  However, there may be situations (e.g., router monitoring)
  where the routing table of only a single router is needed.  In this
  case, the Request should be sent directly to that router from a UDP
  port other than the RIP port.  If such a Request is received, the
  router responds directly to the requestor's address and port.
  The Request is processed entry by entry.  If there are no entries, no
  response is given.  There is one special case.  If there is exactly
  one entry in the request, and it has an address family identifier of
  zero and a metric of infinity (i.e., 16), then this is a request to
  send the entire routing table.  In that case, a call is made to the
  output process to send the routing table to the requesting

Malkin Standards Track [Page 25] � RFC 2453 RIP Version 2 November 1998

  address/port.  Except for this special case, processing is quite
  simple.  Examine the list of RTEs in the Request one by one.  For
  each entry, look up the destination in the router's routing database
  and, if there is a route, put that route's metric in the metric field
  of the RTE.  If there is no explicit route to the specified
  destination, put infinity in the metric field.  Once all the entries
  have been filled in, change the command from Request to Response and
  send the datagram back to the requestor.
  Note that there is a difference in metric handling for specific and
  whole-table requests.  If the request is for a complete routing
  table, normal output processing is done, including Split Horizon (see
  section 3.9 on Split Horizon).  If the request is for specific
  entries, they are looked up in the routing table and the information
  is returned as is; no Split Horizon processing is done.  The reason
  for this distinction is the expectation that these requests are
  likely to be used for different purposes.  When a router first comes
  up, it multicasts a Request on every connected network asking for a
  complete routing table.  It is assumed that these complete routing
  tables are to be used to update the requestor's routing table.  For
  this reason, Split Horizon must be done.  It is further assumed that
  a Request for specific networks is made only by diagnostic software,
  and is not used for routing.  In this case, the requester would want
  to know the exact contents of the routing table and would not want
  any information hidden or modified.

3.9.2 Response Messages

  A Response can be received for one of several different reasons:
  - response to a specific query
  - regular update (unsolicited response)
  - triggered update caused by a route change
  Processing is the same no matter why the Response was generated.
  Because processing of a Response may update the router's routing
  table, the Response must be checked carefully for validity.  The
  Response must be ignored if it is not from the RIP port.  The
  datagram's IPv4 source address should be checked to see whether the
  datagram is from a valid neighbor; the source of the datagram must be
  on a directly-connected network.  It is also worth checking to see
  whether the response is from one of the router's own addresses.
  Interfaces on broadcast networks may receive copies of their own
  broadcasts/multicasts immediately.  If a router processes its own
  output as new input, confusion is likely so such datagrams must be

Malkin Standards Track [Page 26] � RFC 2453 RIP Version 2 November 1998

  Once the datagram as a whole has been validated, process the RTEs in
  the Response one by one.  Again, start by doing validation.
  Incorrect metrics and other format errors usually indicate
  misbehaving neighbors and should probably be brought to the
  administrator's attention.  For example, if the metric is greater
  than infinity, ignore the entry but log the event.  The basic
  validation tests are:
  - is the destination address valid (e.g., unicast; not net 0 or 127)
  - is the metric valid (i.e., between 1 and 16, inclusive)
  If any check fails, ignore that entry and proceed to the next.
  Again, logging the error is probably a good idea.
  Once the entry has been validated, update the metric by adding the
  cost of the network on which the message arrived.  If the result is
  greater than infinity, use infinity.  That is,
  metric = MIN (metric + cost, infinity)
  Now, check to see whether there is already an explicit route for the
  destination address.  If there is no such route, add this route to
  the routing table, unless the metric is infinity (there is no point
  in adding a route which is unusable).  Adding a route to the routing
  table consists of:
  - Setting the destination address to the destination address in the
  - Setting the metric to the newly calculated metric (as described
  - Set the next hop address to be the address of the router from which
    the datagram came
  - Initialize the timeout for the route.  If the garbage-collection
    timer is running for this route, stop it (see section 3.6 for a
    discussion of the timers)
  - Set the route change flag
  - Signal the output process to trigger an update (see section 3.8.1)
  If there is an existing route, compare the next hop address to the
  address of the router from which the datagram came.  If this datagram
  is from the same router as the existing route, reinitialize the
  timeout.  Next, compare the metrics.  If the datagram is from the
  same router as the existing route, and the new metric is different

Malkin Standards Track [Page 27] � RFC 2453 RIP Version 2 November 1998

  than the old one; or, if the new metric is lower than the old one; do
  the following actions:
  - Adopt the route from the datagram (i.e., put the new metric in and
    adjust the next hop address, if necessary).
  - Set the route change flag and signal the output process to trigger
    an update
  - If the new metric is infinity, start the deletion process
    (described above); otherwise, re-initialize the timeout
  If the new metric is infinity, the deletion process begins for the
  route, which is no longer used for routing packets.  Note that the
  deletion process is started only when the metric is first set to
  infinity.  If the metric was already infinity, then a new deletion
  process is not started.
  If the new metric is the same as the old one, it is simplest to do
  nothing further (beyond re-initializing the timeout, as specified
  above); but, there is a heuristic which could be applied.  Normally,
  it is senseless to replace a route if the new route has the same
  metric as the existing route; this would cause the route to bounce
  back and forth, which would generate an intolerable number of
  triggered updates.  However, if the existing route is showing signs
  of timing out, it may be better to switch to an equally-good
  alternative route immediately, rather than waiting for the timeout to
  happen.  Therefore, if the new metric is the same as the old one,
  examine the timeout for the existing route.  If it is at least
  halfway to the expiration point, switch to the new route.  This
  heuristic is optional, but highly recommended.
  Any entry that fails these tests is ignored, as it is no better than
  the current route.

3.10 Output Processing

  This section describes the processing used to create response
  messages that contain all or part of the routing table.  This
  processing may be triggered in any of the following ways:
  - By input processing, when a Request is received (this Response is
    unicast to the requestor; see section 3.7.1)
  - By the regular routing update (broadcast/multicast every 30
    seconds) router.
  - By triggered updates (broadcast/multicast when a route changes)

Malkin Standards Track [Page 28] � RFC 2453 RIP Version 2 November 1998

  When a Response is to be sent to all neighbors (i.e., a regular or
  triggered update), a Response message is directed to the router at
  the far end of each connected point-to-point link, and is broadcast
  (multicast for RIP-2) on all connected networks which support
  broadcasting.  Thus, one Response is prepared for each directly-
  connected network, and sent to the appropriate address (direct or
  broadcast/multicast).  In most cases, this reaches all neighboring
  routers.  However, there are some cases where this may not be good
  enough.  This may involve a network that is not a broadcast network
  (e.g., the ARPANET), or a situation involving dumb routers.  In such
  cases, it may be necessary to specify an actual list of neighboring
  routers and send a datagram to each one explicitly.  It is left to
  the implementor to determine whether such a mechanism is needed, and
  to define how the list is specified.

3.10.1 Triggered Updates

  Triggered updates require special handling for two reasons.  First,
  experience shows that triggered updates can cause excessive load on
  networks with limited capacity or networks with many routers on them.
  Therefore, the protocol requires that implementors include provisions
  to limit the frequency of triggered updates.  After a triggered
  update is sent, a timer should be set for a random interval between 1
  and 5 seconds.  If other changes that would trigger updates occur
  before the timer expires, a single update is triggered when the timer
  expires.  The timer is then reset to another random value between 1
  and 5 seconds.  A triggered update should be suppressed if a regular
  update is due by the time the triggered update would be sent.
  Second, triggered updates do not need to include the entire routing
  table.  In principle, only those routes which have changed need to be
  included.  Therefore, messages generated as part of a triggered
  update must include at least those routes that have their route
  change flag set.  They may include additional routes, at the
  discretion of the implementor; however, sending complete routing
  updates is strongly discouraged.  When a triggered update is
  processed, messages should be generated for every directly-connected
  network.  Split Horizon processing is done when generating triggered
  updates as well as normal updates (see section 3.9).  If, after Split
  Horizon processing for a given network, a changed route will appear
  unchanged on that network (e.g., it appears with an infinite metric),
  the route need not be sent.  If no routes need be sent on that
  network, the update may be omitted.  Once all of the triggered
  updates have been generated, the route change flags should be

Malkin Standards Track [Page 29] � RFC 2453 RIP Version 2 November 1998

  If input processing is allowed while output is being generated,
  appropriate interlocking must be done.  The route change flags should
  not be changed as a result of processing input while a triggered
  update message is being generated.
  The only difference between a triggered update and other update
  messages is the possible omission of routes that have not changed.
  The remaining mechanisms, described in the next section, must be
  applied to all updates.

3.10.2 Generating Response Messages

  This section describes how a Response message is generated for a
  particular directly-connected network:
  Set the version number to either 1 or 2.  The mechanism for deciding
  which version to send is implementation specific; however, if this is
  the Response to a Request, the Response version should match the
  Request version.  Set the command to Response.  Set the bytes labeled
  "must be zero" to zero.  Start filling in RTEs.  Recall that there is
  a limit of 25 RTEs to a Response; if there are more, send the current
  Response and start a new one.  There is no defined limit to the
  number of datagrams which make up a Response.
  To fill in the RTEs, examine each route in the routing table.  If a
  triggered update is being generated, only entries whose route change
  flags are set need be included.  If, after Split Horizon processing,
  the route should not be included, skip it.  If the route is to be
  included, then the destination address and metric are put into the
  RTE.  Routes must be included in the datagram even if their metrics
  are infinite.

Malkin Standards Track [Page 30] � RFC 2453 RIP Version 2 November 1998

4. Protocol Extensions

  This section does not change the RIP protocol per se.  Rather, it
  provides extensions to the message format which allows routers to
  share important additional information.
  The same header format is used for RIP-1 and RIP-2 messages (see
  section 3.4).  The format for the 20-octet route entry (RTE) for
  RIP-2 is:
   0                   1                   2                   3 3
   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
  | Address Family Identifier (2) |        Route Tag (2)          |
  |                         IP Address (4)                        |
  |                         Subnet Mask (4)                       |
  |                         Next Hop (4)                          |
  |                         Metric (4)                            |
  The Address Family Identifier, IP Address, and Metric all have the
  meanings defined in section 3.4.  The Version field will specify
  version number 2 for RIP messages which use authentication or carry
  information in any of the newly defined fields.

4.1 Authentication

  Since authentication is a per message function, and since there is
  only one 2-octet field available in the message header, and since any
  reasonable authentication scheme will require more than two octets,
  the authentication scheme for RIP version 2 will use the space of an
  entire RIP entry.  If the Address Family Identifier of the first (and
  only the first) entry in the message is 0xFFFF, then the remainder of
  the entry contains the authentication.  This means that there can be,
  at most, 24 RIP entries in the remainder of the message.  If
  authentication is not in use, then no entries in the message should
  have an Address Family Identifier of 0xFFFF.  A RIP message which
  contains an authentication entry would begin with the following

Malkin Standards Track [Page 31] � RFC 2453 RIP Version 2 November 1998

   0                   1                   2                   3 3
   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
  | Command (1)   | Version (1)   |            unused             |
  |             0xFFFF            |    Authentication Type (2)    |
  ~                       Authentication (16)                     ~
  Currently, the only Authentication Type is simple password and it is
  type 2.  The remaining 16 octets contain the plain text password.  If
  the password is under 16 octets, it must be left-justified and padded
  to the right with nulls (0x00).

4.2 Route Tag

  The Route Tag (RT) field is an attribute assigned to a route which
  must be preserved and readvertised with a route.  The intended use of
  the Route Tag is to provide a method of separating "internal" RIP
  routes (routes for networks within the RIP routing domain) from
  "external" RIP routes, which may have been imported from an EGP or
  another IGP.
  Routers supporting protocols other than RIP should be configurable to
  allow the Route Tag to be configured for routes imported from
  different sources.  For example, routes imported from EGP or BGP
  should be able to have their Route Tag either set to an arbitrary
  value, or at least to the number of the Autonomous System from which
  the routes were learned.
  Other uses of the Route Tag are valid, as long as all routers in the
  RIP domain use it consistently.  This allows for the possibility of a
  BGP-RIP protocol interactions document, which would describe methods
  for synchronizing routing in a transit network.

4.3 Subnet mask

  The Subnet Mask field contains the subnet mask which is applied to
  the IP address to yield the non-host portion of the address.  If this
  field is zero, then no subnet mask has been included for this entry.
  On an interface where a RIP-1 router may hear and operate on the
  information in a RIP-2 routing entry the following rules apply:
  1) information internal to one network must never be advertised into
     another network,

Malkin Standards Track [Page 32] � RFC 2453 RIP Version 2 November 1998

  2) information about a more specific subnet may not be advertised
     where RIP-1 routers would consider it a host route, and
  3) supernet routes (routes with a netmask less specific than the
     "natural" network mask) must not be advertised where they could be
     misinterpreted by RIP-1 routers.

4.4 Next Hop

  The immediate next hop IP address to which packets to the destination
  specified by this route entry should be forwarded.  Specifying a
  value of in this field indicates that routing should be via
  the originator of the RIP advertisement.  An address specified as a
  next hop must, per force, be directly reachable on the logical subnet
  over which the advertisement is made.
  The purpose of the Next Hop field is to eliminate packets being
  routed through extra hops in the system.  It is particularly useful
  when RIP is not being run on all of the routers on a network.  A
  simple example is given in Appendix A.  Note that Next Hop is an
  "advisory" field.  That is, if the provided information is ignored, a
  possibly sub-optimal, but absolutely valid, route may be taken.  If
  the received Next Hop is not directly reachable, it should be treated

4.5 Multicasting

  In order to reduce unnecessary load on those hosts which are not
  listening to RIP-2 messages, an IP multicast address will be used for
  periodic broadcasts.  The IP multicast address is  Note
  that IGMP is not needed since these are inter-router messages which
  are not forwarded.
  On NBMA networks, unicast addressing may be used.  However, if a
  response addressed to the RIP-2 multicast address is received, it
  should be accepted.
  In order to maintain backwards compatibility, the use of the
  multicast address will be configurable, as described in section 5.1.
  If multicasting is used, it should be used on all interfaces which
  support it.

4.6 Queries

  If a RIP-2 router receives a RIP-1 Request, it should respond with a
  RIP-1 Response.  If the router is configured to send only RIP-2
  messages, it should not respond to a RIP-1 Request.

Malkin Standards Track [Page 33] � RFC 2453 RIP Version 2 November 1998

5. Compatibility

  RFC [1] showed considerable forethought in its specification of the
  handling of version numbers.  It specifies that RIP messages of
  version 0 are to be discarded, that RIP messages of version 1 are to
  be discarded if any Must Be Zero (MBZ) field is non-zero, and that
  RIP messages of any version greater than 1 should not be discarded
  simply because an MBZ field contains a value other than zero.  This
  means that the new version of RIP is totally backwards compatible
  with existing RIP implementations which adhere to this part of the

5.1 Compatibility Switch

  A compatibility switch is necessary for two reasons.  First, there
  are implementations of RIP-1 in the field which do not follow RFC [1]
  as described above.  Second, the use of multicasting would prevent
  RIP-1 systems from receiving RIP-2 updates (which may be a desired
  feature in some cases).  This switch should be configurable on a
  per-interface basis.
  The switch has four settings: RIP-1, in which only RIP-1 messages are
  sent; RIP-1 compatibility, in which RIP-2 messages are broadcast;
  RIP-2, in which RIP-2 messages are multicast; and "none", which
  disables the sending of RIP messages.  It is recommended that the
  default setting be either RIP-1 or RIP-2, but not RIP-1
  compatibility.  This is because of the potential problems which can
  occur on some topologies.  RIP-1 compatibility should only be used
  when all of the consequences of its use are well understood by the
  network administrator.
  For completeness, routers should also implement a receive control
  switch which would determine whether to accept, RIP-1 only, RIP-2
  only, both, or none.  It should also be configurable on a per-
  interface basis.  It is recommended that the default be compatible
  with the default chosen for sending updates.

5.2 Authentication

  The following algorithm should be used to authenticate a RIP message.
  If the router is not configured to authenticate RIP-2 messages, then
  RIP-1 and unauthenticated RIP-2 messages will be accepted;
  authenticated RIP-2 messages shall be discarded.  If the router is
  configured to authenticate RIP-2 messages, then RIP-1 messages and
  RIP-2 messages which pass authentication testing shall be accepted;
  unauthenticated and failed authentication RIP-2 messages shall be
  discarded.  For maximum security, RIP-1 messages should be ignored

Malkin Standards Track [Page 34] � RFC 2453 RIP Version 2 November 1998

  when authentication is in use (see section 4.1); otherwise, the
  routing information from authenticated messages will be propagated by
  RIP-1 routers in an unauthenticated manner.
  Since an authentication entry is marked with an Address Family
  Identifier of 0xFFFF, a RIP-1 system would ignore this entry since it
  would belong to an address family other than IP.  It should be noted,
  therefore, that use of authentication will not prevent RIP-1 systems
  from seeing RIP-2 messages.  If desired, this may be done using
  multicasting, as described in sections 4.5 and 5.1.

5.3 Larger Infinity

  While on the subject of compatibility, there is one item which people
  have requested: increasing infinity.  The primary reason that this
  cannot be done is that it would violate backwards compatibility.  A
  larger infinity would obviously confuse older versions of rip.  At
  best, they would ignore the route as they would ignore a metric of
  16.  There was also a proposal to make the Metric a single octet and
  reuse the high three octets, but this would break any implementations
  which treat the metric as a 4-octet entity.

5.4 Addressless Links

  As in RIP-1, addressless links will not be supported by RIP-2.

6. Interaction between version 1 and version 2

  Because version 1 packets do not contain subnet information, the
  semantics employed by routers on networks that contain both version 1
  and version 2 networks should be limited to that of version 1.
  Otherwise it is possible either to create blackhole routes (i.e.,
  routes for networks that do not exist) or to create excessive routing
  information in a version 1 environment.
  Some implementations attempt to automatically summarize groups of
  adjacent routes into single entries, the goal being to reduce the
  total number of entries.  This is called auto-summarization.
  Specifically, when using both version 1 and version 2 within a
  network, a single subnet mask should be used throughout the network.
  In addition, auto-summarization mechanisms should be disabled for
  such networks, and implementations must provide mechanisms to disable

Malkin Standards Track [Page 35] � RFC 2453 RIP Version 2 November 1998

7. Security Considerations

  The basic RIP protocol is not a secure protocol.  To bring RIP-2 in
  line with more modern routing protocols, an extensible authentication
  mechanism has been incorporated into the protocol enhancements.  This
  mechanism is described in sections 4.1 and 5.2.  Security is further
  enhanced by the mechanism described in [3].

Malkin Standards Track [Page 36] � RFC 2453 RIP Version 2 November 1998

Appendix A

  This is a simple example of the use of the next hop field in a rip
     -----   -----   -----           -----   -----   -----
     |IR1|   |IR2|   |IR3|           |XR1|   |XR2|   |XR3|
     --+--   --+--   --+--           --+--   --+--   --+--
       |       |       |               |       |       |
  Assume that IR1, IR2, and IR3 are all "internal" routers which are
  under one administration (e.g. a campus) which has elected to use
  RIP-2 as its IGP. XR1, XR2, and XR3, on the other hand, are under
  separate administration (e.g. a regional network, of which the campus
  is a member) and are using some other routing protocol (e.g. OSPF).
  XR1, XR2, and XR3 exchange routing information among themselves such
  that they know that the best routes to networks N1 and N2 are via
  XR1, to N3, N4, and N5 are via XR2, and to N6 and N7 are via XR3. By
  setting the Next Hop field correctly (to XR2 for N3/N4/N5, to XR3 for
  N6/N7), only XR1 need exchange RIP-2 routes with IR1/IR2/IR3 for
  routing to occur without additional hops through XR1. Without the
  Next Hop (for example, if RIP-1 were used) it would be necessary for
  XR2 and XR3 to also participate in the RIP-2 protocol to eliminate
  extra hops.


  [1] Hedrick, C., "Routing Information Protocol", STD 34, RFC 1058,
      Rutgers  University, June 1988.
  [2] Malkin, G., and F. Baker, "RIP Version 2 MIB Extension", RFC 1389, January 1993.
  [3] Baker, F., and R. Atkinson, "RIP-II MD5 Authentication", RFC 2082, January 1997.
  [4] Bellman, R. E., "Dynamic Programming", Princeton University
      Press, Princeton, N.J., 1957.
  [5] Bertsekas, D. P., and Gallaher, R. G., "Data Networks",
      Prentice-Hall, Englewood Cliffs, N.J., 1987.
  [6] Braden, R., and Postel, J., "Requirements for Internet Gateways",
      STD 4, RFC 1009, June 1987.

Malkin Standards Track [Page 37] � RFC 2453 RIP Version 2 November 1998

  [7] Boggs, D. R., Shoch, J. F., Taft, E. A., and Metcalfe, R. M.,
      "Pup: An Internetwork Architecture", IEEE Transactions on
      Communications, April 1980.
  [8] Ford, L. R. Jr., and Fulkerson, D. R., "Flows in Networks",
      Princeton University Press, Princeton, N.J., 1962.
  [9] Xerox Corp., "Internet Transport Protocols", Xerox System
      Integration Standard XSIS 028112, December 1981.
  [10] Floyd, S., and V. Jacobson, "The synchronization of Periodic
       Routing Messages," ACM Sigcom '93 symposium, September 1993.
  [11] Baker, F., "Requirements for IP Version 4 Routers." RFC 1812,
       June 1995.

Author's Address

  Gary Scott Malkin
  Bay Networks
  8 Federal Street
  Billerica, MA 01821
  Phone:  (978) 916-4237

Malkin Standards Track [Page 38] � RFC 2453 RIP Version 2 November 1998

Full Copyright Statement

  Copyright (C) The Internet Society (1998).  All Rights Reserved.
  This document and translations of it may be copied and furnished to
  others, and derivative works that comment on or otherwise explain it
  or assist in its implementation may be prepared, copied, published
  and distributed, in whole or in part, without restriction of any
  kind, provided that the above copyright notice and this paragraph are
  included on all such copies and derivative works.  However, this
  document itself may not be modified in any way, such as by removing
  the copyright notice or references to the Internet Society or other
  Internet organizations, except as needed for the purpose of
  developing Internet standards in which case the procedures for
  copyrights defined in the Internet Standards process must be
  followed, or as required to translate it into languages other than
  The limited permissions granted above are perpetual and will not be
  revoked by the Internet Society or its successors or assigns.
  This document and the information contained herein is provided on an

Malkin Standards Track [Page 39] �