Home
You are not currently signed in.

RFC9494

  1. RFC 9494
Internet Engineering Task Force (IETF)                         J. Uttaro
Request for Comments: 9494                       Independent Contributor
Updates: 6368                                                    E. Chen
Category: Standards Track                             Palo Alto Networks
ISSN: 2070-1721                                              B. Decraene
                                                                  Orange
                                                              J. Scudder
                                                        Juniper Networks
                                                           November 2023


                  Long-Lived Graceful Restart for BGP

Abstract

   This document introduces a BGP capability called the "Long-Lived
   Graceful Restart Capability" (or "LLGR Capability").  The benefit of
   this capability is that stale routes can be retained for a longer
   time upon session failure than is provided for by BGP Graceful
   Restart (as described in RFC 4724).  A well-known BGP community
   called "LLGR_STALE" is introduced for marking stale routes retained
   for a longer time.  A second well-known BGP community called
   "NO_LLGR" is introduced for marking routes for which these procedures
   should not be applied.  We also specify that such long-lived stale
   routes be treated as the least preferred and that their
   advertisements be limited to BGP speakers that have advertised the
   capability.  Use of this extension is not advisable in all cases, and
   we provide guidelines to help determine if it is.

   This memo updates RFC 6368 by specifying that the LLGR_STALE
   community must be propagated into, or out of, the path attributes
   exchanged between the Provider Edge (PE) and Customer Edge (CE)
   routers.

Status of This Memo

   This is an Internet Standards Track document.

   This document is a product of the Internet Engineering Task Force
   (IETF).  It represents the consensus of the IETF community.  It has
   received public review and has been approved for publication by the
   Internet Engineering Steering Group (IESG).  Further information on
   Internet Standards is available in Section 2 of RFC 7841.

   Information about the current status of this document, any errata,
   and how to provide feedback on it may be obtained at
   https://www.rfc-editor.org/info/rfc9494.

Copyright Notice

   Copyright (c) 2023 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (https://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Revised BSD License text as described in Section 4.e of the
   Trust Legal Provisions and are provided without warranty as described
   in the Revised BSD License.

Table of Contents

   1.  Introduction
   2.  Terminology
     2.1.  Definitions
     2.2.  Abbreviations
     2.3.  Requirements Language
   3.  Protocol Extensions
     3.1.  Long-Lived Graceful Restart Capability
     3.2.  LLGR_STALE Community
     3.3.  NO_LLGR Community
   4.  Theory of Operation
     4.1.  Use of the Graceful Restart Capability
     4.2.  Session Resets
     4.3.  Processing LLGR_STALE Routes
     4.4.  Route Selection
     4.5.  Errors
     4.6.  Optional Partial Deployment Procedure
     4.7.  Procedures When BGP Is the PE-CE Protocol in a VPN
       4.7.1.  Procedures When EBGP Is the PE-CE Protocol in a VPN
       4.7.2.  Procedures When IBGP Is the PE-CE Protocol in a VPN
   5.  Deployment Considerations
     5.1.  When BGP Is the PE-CE Protocol in a VPN
     5.2.  Risks of Depreferencing Routes
   6.  Security Considerations
   7.  Examples of Operation
   8.  IANA Considerations
   9.  References
     9.1.  Normative References
     9.2.  Informative References
   Acknowledgements
   Contributors
   Authors' Addresses

1.  Introduction

   Routing protocols in general, and BGP in particular, have
   historically been designed with a focus on "correctness", where a key
   part of correctness is for each network element's forwarding state to
   converge to the current state of the network as quickly as possible.
   For this reason, the protocol was designed to remove state advertised
   by routers that went down (from a BGP perspective) as quickly as
   possible.  Over time, this has been relaxed somewhat, notably by BGP
   Graceful Restart (GR) [RFC4724]; however, the paradigm has remained
   one of attempting to rapidly remove stale state from the network.

   Over time, two phenomena have arisen that call into question the
   underlying assumptions of this paradigm.

   1.  The widespread adoption of tunneled forwarding infrastructures
       (for example, MPLS).  Such infrastructures eliminate the risk of
       some types of forwarding loops that can arise in hop-by-hop
       forwarding; thus, they reduce one of the motivations for strong
       consistency between forwarding elements.

   2.  The increasing use of BGP as a transport for data that is less
       closely associated with packet forwarding than was originally the
       case.  Examples include the use of BGP for auto-discovery
       (Virtual Private LAN Service (VPLS) [RFC4761]) and filter
       programming (Flow Specification (FLOWSPEC) [RFC8955]).  In these
       cases, BGP data takes on a character more akin to configuration
       than to conventional routing.

   The observations above motivate a desire to offer network operators
   the ability to choose to retain BGP data for a longer period than has
   hitherto been possible when the BGP control plane fails for some
   reason.  Although the semantics of BGP Graceful Restart [RFC4724] are
   close to those desired, several gaps exist, most notably in the
   maximum time for which stale information can be retained: Graceful
   Restart imposes a 4095-second upper bound.

   In this document, we introduce a BGP capability called the "Long-
   Lived Graceful Restart Capability".  The goal of this capability is
   that stale information can be retained for a longer time across a
   session reset.  We also introduce two BGP well-known communities:

   *  LLGR_STALE to mark such information, and

   *  NO_LLGR to indicate that these procedures should not be applied to
      the marked route.

   Long-lived stale information is to be treated as least preferred, and
   its advertisement limited to BGP speakers that support the
   capability.  Where possible, we reference the semantics of BGP
   Graceful Restart [RFC4724] rather than specifying similar semantics
   in this document.

   The expected deployment model for this extension is that it will only
   be invoked for certain address families.  This is discussed in more
   detail in Section 5.  The use of this extension may be combined with
   that of conventional Graceful Restart; in such a case, it is invoked
   after the conventional Graceful Restart interval has elapsed.  When
   not combined, LLGR is invoked immediately.  Apart from the potential
   to greatly extend the timer, the most obvious difference between LLGR
   and conventional Graceful Restart is that in LLGR, routes are
   "depreferenced"; that is, they are treated as least preferred.
   Contrarily, in conventional GR, route preference is not affected.
   The design choice to treat long-lived stale routes as least preferred
   was informed by the expectation that they might be retained for
   (potentially) an almost unbounded period of time; whereas, in the
   conventional Graceful Restart case, stale routes are retained for
   only a brief interval.  In the case of Graceful Restart, the trade-
   off between advertising new route status (at the cost of routing
   churn) and not advertising it (at the cost of suboptimal or incorrect
   route selection) is resolved in favor of not advertising.  In the
   case of LLGR, it is resolved in favor of advertising new state, using
   stale information only as a last resort.

   Section 7 provides some simple examples illustrating the operation of
   this extension.

2.  Terminology

2.1.  Definitions

   Depreference:  A route is said to be depreferenced if it has its
     route selection preference reduced in reaction to some event.

   Helper:  Sometimes referred to as "helper router".  During Graceful
     Restart or Long-Lived Graceful Restart, the router that detects a
     session failure and applies the listed procedures.  [RFC4724]
     refers to this as the "receiving speaker".

   Route:  In this document, "route" means any information encoded as
     BGP Network Layer Reachability Information (NLRI) and a set of path
     attributes.  As discussed above, the connection between such routes
     and the installation of forwarding state may be quite remote.

   Further note that, for brevity, in this document when we reference
   conventional Graceful Restart, we cite its base specification,
   [RFC4724].  That specification has been updated by [RFC8538].  The
   citation to [RFC4724] is not intended to be limiting.

2.2.  Abbreviations

   CE:  Customer Edge (See [RFC4364] for more information on Customer
     Edge routers.)

   EoR:  End-of-RIB (See Section 2 of [RFC4724] for more information on
     End-of-RIB markers.)

   GR:  Graceful Restart (See [RFC4724] for more information on GR.)
     This term is also sometimes referred to herein as "conventional
     Graceful Restart" or "conventional GR" to distinguish it from the
     "Long-Lived Graceful Restart" or "LLGR" defined by this document.

   LLGR:  Long-Lived Graceful Restart

   LLST:  Long-Lived Stale Time

   PE:  Provider Edge (See [RFC4364] for more information on Provider
     Edge routers.)

   VRF:  VPN Routing and Forwarding (See [RFC4364] for more information
     on VRF tables.)

2.3.  Requirements Language

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
   "OPTIONAL" in this document are to be interpreted as described in
   BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all
   capitals, as shown here.

3.  Protocol Extensions

   A BGP capability and two BGP communities are introduced in the
   subsections that follow.

3.1.  Long-Lived Graceful Restart Capability

   The "Long-Lived Graceful Restart Capability", or "LLGR Capability",
   (value: 71) is a BGP capability [RFC5492] that can be used by a BGP
   speaker to indicate its ability to preserve its state according to
   the procedures of this document.  If the LLGR capability is
   advertised, the Graceful Restart capability [RFC4724] MUST also be
   advertised; see Section 4.1.

   The capability value consists of zero or more tuples <AFI, SAFI,
   Flags, LLST> as follows:

   +--------------------------------------------------+
   | Address Family Identifier (16 bits)              |
   +--------------------------------------------------+
   | Subsequent Address Family Identifier (8 bits)    |
   +--------------------------------------------------+
   | Flags for Address Family (8 bits)                |
   +--------------------------------------------------+
   | Long-Lived Stale Time (24 bits)                  |
   +--------------------------------------------------+
   | ...                                              |
   +--------------------------------------------------+
   | Address Family Identifier (16 bits)              |
   +--------------------------------------------------+
   | Subsequent Address Family Identifier (8 bits)    |
   +--------------------------------------------------+
   | Flags for Address Family (8 bits)                |
   +--------------------------------------------------+
   | Long-Lived Stale Time (24 bits)                  |
   +--------------------------------------------------+

   The meaning of the fields are as follows:

   Address Family Identifier (AFI), Subsequent Address Family
   Identifier (SAFI):
      The AFI and SAFI, taken in combination, indicate that the BGP
      speaker has the ability to preserve its forwarding state for the
      address family during a subsequent BGP restart.  Routes may be
      either:

      *  explicitly associated with a particular AFI and SAFI if using
         the encoding described in [RFC4760], or

      *  implicitly associated with <AFI=IPv4, SAFI=Unicast> if using
         the encoding described in [RFC4271].

   Flags for Address Family:
      This field contains bit flags relating to routes that were
      advertised with the given AFI and SAFI.

                              0 1 2 3 4 5 6 7
                             +-+-+-+-+-+-+-+-+
                             |F|   Reserved  |
                             +-+-+-+-+-+-+-+-+

      The most significant bit is used to indicate whether the state for
      routes that were advertised with the given AFI and SAFI has indeed
      been preserved during the previous BGP restart.  When set (value
      1), the bit indicates that the state has been preserved.  This bit
      is called the "F bit" since it was historically used to indicate
      the preservation of forwarding state.  Use of the F bit is
      detailed in Section 4.2.  The remaining bits are reserved and MUST
      be set to zero by the sender and ignored by the receiver.

   Long-Lived Stale Time:
      This time (in seconds) specifies how long stale information (for
      this AFI/SAFI) may be retained by the receiver (in addition to the
      period specified by the "Restart Time" in the Graceful Restart
      Capability).  Because the potential use cases for this extension
      vary widely, there is no suggested default value for the LLST.

3.2.  LLGR_STALE Community

   The well-known BGP community LLGR_STALE (value: 0xFFFF0006) can be
   used to mark stale routes retained for a longer period of time (see
   [RFC1997] for more information on BGP communities).  Such long-lived
   stale routes are to be handled according to the procedures specified
   in Section 4.

   An implementation MAY allow users to configure policies that accept,
   reject, or modify routes based on the presence or absence of this
   community.

3.3.  NO_LLGR Community

   The well-known BGP community NO_LLGR (value: 0xFFFF0007) can be used
   to mark routes that a BGP speaker does not want to be treated
   according to these procedures, as detailed in Section 4.

   An implementation MAY allow users to configure policies that accept,
   reject, or modify routes based on the presence or absence of this
   community.

4.  Theory of Operation

   If a BGP speaker is configured to support the procedures of this
   document, it MUST use BGP Capabilities Advertisement [RFC5492] to
   advertise the Long-Lived Graceful Restart Capability.  The setting of
   the parameters for an AFI/SAFI depends on the properties of the BGP
   speaker, network scale, and local configuration.

   In the presence of the Long-Lived Graceful Restart Capability, the
   procedures specified in [RFC4724] continue to apply unless explicitly
   revised by this document.

4.1.  Use of the Graceful Restart Capability

   If the LLGR Capability is advertised, the Graceful Restart capability
   MUST also be advertised.  If it is not so advertised, the LLGR
   Capability MUST be disregarded.  The purpose for mandating this is to
   enable the reuse of certain base mechanisms that are common to both
   "flavors" notably: origination, collection, and processing of EoR as
   well as the finite-state-machine modifications and connection-reset
   logic introduced by GR.

   We observe that, if support for conventional Graceful Restart is not
   desired for the session, the conventional GR phase can be skipped by
   omitting all AFIs/SAFIs from the GR Capability, advertising a Restart
   Time of zero, or both.  Section 4.2 discusses the interaction of
   conventional and LLGR.

4.2.  Session Resets

   BGP Graceful Restart [RFC4724] defines conditions under which a BGP
   session can reset and have its associated routes retained.  If such a
   reset occurs for a session in which the LLGR Capability has also been
   exchanged, the following procedures apply:

   *  If the Graceful Restart Capability that was received does not list
      all AFIs/SAFIs supported by the session, then the GR Restart Time
      shall be deemed zero for those AFIs/SAFIs that are not listed.

   *  Similarly, if the received LLGR Capability does not list all AFIs/
      SAFIs supported by the session, then the Long-Lived Stale Time
      shall be deemed zero for those AFIs/SAFIs that are not listed.

   The following text in Section 4.2 of [RFC4724] no longer applies:

   |  If the session does not get re-established within the "Restart
   |  Time" that the peer advertised previously, the Receiving Speaker
   |  MUST delete all the stale routes from the peer that it is
   |  retaining.

   and the following procedures are specified instead:

   After the session goes down, and before the session is re-
   established, the stale routes for an AFI/SAFI MUST be retained.  The
   interval for which they are retained is limited by the sum of the
   Restart Time in the received Graceful Restart Capability and the
   Long-Lived Stale Time in the received Long-Lived Graceful Restart
   Capability.  The timers received in the Long-Lived Graceful Restart
   Capability SHOULD be modifiable by local configuration, which may
   impose an upper bound, a lower bound, or both on their respective
   values.

   If the value of the Restart Time or the Long-Lived Stale Time is
   zero, the duration of the corresponding period would be zero seconds.
   For example, if the Restart Time is zero and the Long-Lived Stale
   Time is nonzero, only the procedures particular to LLGR would apply.
   Conversely, if the Long-Lived Stale Time is zero and the Restart Time
   is nonzero, only the procedures of GR would apply.  If both are zero,
   none of these procedures would apply, only those of the base BGP
   specification [RFC4271] (although EoR would still be used as detailed
   in [RFC4724]).  And finally, if both are nonzero, then the procedures
   would be applied serially: first those of GR and then those of LLGR.
   During the first interval, we observe that, while the procedures of
   GR are in effect, route preference would not be affected.  During the
   second interval, while LLGR procedures are in effect, routes would be
   treated as least preferred as specified elsewhere in this document.

   Once the Restart Time period ends (including the case in which the
   Restart Time is zero), the LLGR period is said to have begun and the
   following procedures MUST be performed:

   *  For each AFI/SAFI for which it has received a nonzero Long-Lived
      Stale Time, the helper router MUST start a timer for that Long-
      Lived Stale Time.  If the timer for the Long-Lived Stale Time for
      a given AFI/SAFI expires before the session is re-established, the
      helper MUST delete all stale routes of that AFI/SAFI from the
      neighbor that it is retaining.

   *  The helper router MUST attach the LLGR_STALE community to the
      stale routes being retained.  Note that this requirement implies
      that the routes would need to be readvertised in order to
      disseminate the modified community.

   *  If any of the routes from the peer have been marked with the
      NO_LLGR community, either as sent by the peer or as the result of
      a configured policy, they MUST NOT be retained and MUST be removed
      as per the normal operation of [RFC4271].

   *  The helper router MUST perform the procedures listed in
      Section 4.3.

   Once the session is re-established, the procedures specified in
   [RFC4724] apply for the stale routes irrespective of whether the
   stale routes are retained during the Restart Time period or the Long-
   Lived Stale Time period.  However, in the case of consecutive
   restarts, the previously marked stale routes MUST NOT be deleted
   before the timer for the Long-Lived Stale Time expires.

   Similar to [RFC4724], once the LLGR Period begins, the Helper MUST
   immediately remove all the stale routes from the peer that it is
   retaining for that address family if any of the following occur:

   *  the F bit for a specific address family is not set in the newly
      received LLGR Capability, or

   *  a specific address family is not included in the newly received
      LLGR Capability, or

   *  the LLGR and accompanying GR Capability are not received in the
      re-established session at all.

   If a Long-Lived Stale Time timer is running for routes with a given
   AFI/SAFI received from a peer, it MUST NOT be updated (other than by
   manual operator intervention) until the peer has established and
   synchronized a new session.  The session is termed "synchronized" for
   a given AFI/SAFI once the EoR for that AFI/SAFI has been received
   from the peer or once the Selection_Deferral_Timer discussed in
   [RFC4724] expires.

   The value of a Long-Lived Stale Time in the capability received from
   a neighbor MAY be reduced by local configuration.

   While the session is down, the expiration of a Long-Lived Stale Time
   timer is treated analogously to the expiration of the Restart Time
   timer in [RFC4724], other than applying only to the AFI/SAFI it
   accompanies.  However, the timer continues to run once the session
   has re-established.  The timer is neither stopped nor updated until
   the EoR marker is received for the relevant AFI/SAFI from the peer.
   If the timer expires during synchronization with the peer, any stale
   routes that the peer has not refreshed are removed.  If the session
   subsequently resets prior to becoming synchronized, any remaining
   routes (for the AFI/SAFI whose LLST timer expired) MUST be removed
   immediately.

4.3.  Processing LLGR_STALE Routes

   A BGP speaker that has advertised the Long-Lived Graceful Restart
   Capability to a neighbor MUST perform the following upon receiving a
   route from that neighbor with the LLGR_STALE community or upon
   attaching the LLGR_STALE community itself per Section 4.2:

   *  Treat the route as the least preferred in route selection (see
      below).  See Section 5.2 for a discussion of potential risks
      inherent in doing this.

   *  The route SHOULD NOT be advertised to any neighbor from which the
      Long-Lived Graceful Restart Capability has not been received.  The
      exception is described in Section 4.6.  Note that this requirement
      implies that such routes should be withdrawn from any such
      neighbor.

   *  The LLGR_STALE community MUST NOT be removed when the route is
      further advertised.

4.4.  Route Selection

   A least preferred route MUST be treated as less preferred than any
   other route that is not also least preferred.  When performing route
   selection between two routes when both are least preferred, normal
   tiebreaking applies.  Note that this would only be expected to happen
   if the only routes available for selection were least preferred; in
   all other cases, such routes would have been eliminated from
   consideration.

4.5.  Errors

   If the LLGR Capability is received without an accompanying GR
   Capability, the LLGR Capability MUST be ignored, that is, the
   implementation MUST behave as though no LLGR Capability has been
   received.

4.6.  Optional Partial Deployment Procedure

   Ideally, all routers in an Autonomous System (AS) would support this
   specification before it were enabled.  However, to facilitate
   incremental deployment, stale routes MAY be advertised to neighbors
   that have not advertised the Long-Lived Graceful Restart Capability
   under the following conditions:

   *  The neighbors MUST be internal (Internal BGP (IBGP) or
      Confederation) neighbors.

   *  The NO_EXPORT community [RFC1997] MUST be attached to the stale
      routes.

   *  The stale routes MUST have their LOCAL_PREF set to zero.  See
      Section 5.2 for a discussion of potential risks inherent in doing
      this.

   If this strategy for partial deployment is used, the network operator
   should set the LOCAL_PREF to zero for all long-lived stale routes
   throughout the Autonomous System.  This trades off a small reduction
   in flexibility (ordering may not be preserved between competing long-
   lived stale routes) for consistency between routers that do, and do
   not, support this specification.  Since the consistency of route
   selection can be important for preventing forwarding loops, the
   latter consideration dominates.

4.7.  Procedures When BGP Is the PE-CE Protocol in a VPN

4.7.1.  Procedures When EBGP Is the PE-CE Protocol in a VPN

   In VPN deployments (for example, [RFC4364]), External BGP (EBGP) is
   often used as a PE-CE protocol.  It may be a practical necessity in
   such deployments to accommodate interoperation with peer routers that
   cannot easily be upgraded to support specifications such as this one.
   This leads to a problem: the procedures defined elsewhere in this
   document generally prevent LLGR stale routes from being sent across
   EBGP sessions that don't support LLGR, but this could prevent the VPN
   routes from being used for their intended purpose.

   We observe that the principal motivation for restricting the
   propagation of "stale" routing information is the desire to prevent
   it from spreading without limit once it exits the "safe" perimeter.
   We further observe that VPN deployments are typically topologically
   constrained, making this concern moot.  For this reason, an
   implementation MAY advertise stale routes over a PE-CE session, when
   explicitly configured to do so.  That is, the second rule listed in
   Section 4.3 MAY be disregarded in such cases.  All other rules
   continue to apply.  Finally, if this exception is used, the
   implementation SHOULD, by default, attach the NO_EXPORT community to
   the routes in question, as an additional protection against stale
   routes spreading without limit.  Attachment of the NO_EXPORT
   community MAY be disabled by explicit configuration in order to
   accommodate exceptional cases.

   See further discussion of using an explicitly configured policy to
   mitigate this issue in Section 5.1.

4.7.2.  Procedures When IBGP Is the PE-CE Protocol in a VPN

   If IBGP is used as the PE-CE protocol, following the procedures of
   [RFC6368], then when a PE router imports a VPN route that contains
   the ATTR_SET attribute into a destination VRF and subsequently
   advertises that route to a CE router:

   *  If the CE router supports the procedures of this document (in
      other words, if the CE router has advertised the LLGR Capability):

         In addition to including the path attributes derived from the
         ATTR_SET attribute in the advertised route as per [RFC6368],
         the PE router MUST also include the LLGR_STALE community if it
         is present in the path attributes of the imported route, even
         if it is not present in the ATTR_SET attribute.

   *  If the CE router does not support the procedures of this document:

         Then the optional procedures of Section 4.6 MAY be followed,
         attaching the NO_EXPORT community and setting the value of
         LOCAL_PREF to zero, overriding the value found in the ATTR_SET.

   Similarly, when a PE router receives a route from a CE into its VRF
   and subsequently exports that route to a VPN address family:

   *  If the PE router supports the procedures of this document (in
      other words, if the PE router has advertised the LLGR Capability):

         In addition to including in the VPN route the ATTR_SET derived
         from the path attributes as per [RFC6368], the PE router MUST
         also include the LLGR_STALE community in the VPN route if it is
         present in the path attributes of the route as received from
         the CE.

   *  If the PE router does not support the procedures of this document:

         There exists no ideal solution.  The CE could advertise a route
         with LLGR_STALE, with the understanding that the LLGR_STALE
         marking will only be honored by the provider network if
         appropriate policy configuration exists on the PE (see
         Section 5.1).  It is at least guaranteed that LLGR_STALE will
         be propagated when the route is propagated beyond the provider
         network, or the CE could refrain from advertising the
         LLGR_STALE route to the incapable PE.

5.  Deployment Considerations

   The deployment considerations discussed in [RFC4724] apply to this
   document.  In addition, network operators are cautioned to carefully
   consider the potential disadvantages of deploying these procedures
   for a given AFI/SAFI.  Most notably, if used for an AFI/SAFI that
   conveys conventional reachability information, the use of a long-
   lived stale route could result in a loss of connectivity for the
   covered prefix.  This specification takes pains to mitigate this risk
   where possible by making such routes least preferred and by
   restricting the scope of such routes to routers that support these
   procedures (or, optionally, a single Autonomous System, see
   Section 4.6).  However, if a stale route is chosen as best for a
   given prefix, then according to the normal rules of IP forwarding,
   that route will be used for matching destinations, even if a non-
   stale less specific matching route is also available.  Networks in
   which the deployment of these procedures would be especially
   concerning include those that do not use "tunneled" forwarding (in
   other words, those using conventional hop-by-hop forwarding).

   Implementations MUST NOT enable these procedures by default.  They
   MUST require affirmative configuration per AFI/SAFI in order to
   enable them.

   The procedures of this document do not alter the route resolvability
   requirement of Section 9.1.2.1 of [RFC4271].  Because of this, it
   will commonly be the case that "stale" IBGP routes will only continue
   to be used if the router depicted in the next hop remains resolvable,
   even if its BGP component is down.  Details of IGP fault-tolerance
   strategies are beyond the scope of this document.  In addition to the
   foregoing, it may be advisable to check the viability of the next hop
   through other means, for example, Bidirectional Forwarding Detection
   (BFD) [RFC5880].  This may be especially useful in cases where the
   next hop is known directly at the network layer, notably EBGP.

   As discussed in this document, after a BGP session goes down and
   before the session is re-established, stale routes may be retained
   for up to two consecutive periods, controlled by the Restart Time and
   the Long-Lived Stale Time, respectively:

   *  During the first period, routing churn would be prevented, but
      with potential persistent packet loss.

   *  During the second period, potential persistent packet loss may be
      reduced, but routing churn would be visible throughout the
      network.

   The setting of the relevant parameters for a particular application
   should take into account trade-offs, network dynamics, and potential
   failure scenarios.  If needed, the first period can be bypassed
   either by local configuration or by setting the Restart Time in the
   Graceful Restart Capability to zero and/or not listing the AFI/SAFI
   in that capability.

   The setting of the F bit (and the Forwarding State bit of the
   accompanying GR Capability) depends, in part, on deployment
   considerations.  The F bit can be understood as an indication that
   the Helper should flush associated routes (if the bit is left clear).
   As discussed in Section 1, an important use case for LLGR is for
   routes that are more akin to configuration than to conventional
   routing.  For such routes, it may make sense to always set the F bit,
   regardless of other considerations.  Likewise, for control-plane-only
   entities, such as dedicated route reflectors that do not participate
   in the forwarding plane, it makes sense to always set the F bit.
   Overall, the rule of thumb is that if loss of state on the restarting
   router can reasonably be expected to cause a forwarding loop or
   persistent packet loss, the F bit should be set scrupulously
   according to whether state has been retained.  Specifics of whether
   or not the F bit is set are implementation dependent and may also be
   controlled by configuration.  Also, for every AFI/SAFI represented in
   the LLGR Capability that is also represented in the GR Capability,
   there will be two corresponding F bits: the LLGR F bit and the GR F
   bit.  If the LLGR F bit is set, the corresponding GR F bit should
   also be set, since to do otherwise would cause the state to be
   cleared on the Receiving Router per the normal rules of GR, violating
   the intent of the set LLGR bit.

5.1.  When BGP Is the PE-CE Protocol in a VPN

   As discussed in Section 4.7, it may be necessary for a PE to
   advertise stale routes to a CE in some VPN deployments, even if the
   CE does not support this specification.  In that case, the operator
   configuring their PE to advertise such routes should notify the
   operator of the CE receiving the routes, and the CE should be
   configured to depreference the routes.

   Similarly, it may be necessary for a CE to advertise stale routes to
   a PE, even if the PE does not support this specification.  In that
   case, the operator configuring their CE to advertise such routes
   should notify the operator of the PE receiving the routes, and the PE
   should be configured to depreference the routes.

   Typical BGP implementations will be able to be configured to
   depreference routes by matching on the LLGR_STALE community and
   setting the LOCAL_PREF for matching routes to zero, similar to the
   procedure described in Section 4.6.

5.2.  Risks of Depreferencing Routes

   Depreferencing EBGP routes is considered safe, no different from the
   common practice of applying a routing policy to an EBGP session.
   However, the same is not always true of IBGP.

   Consistent route selection is a fundamental tenet of IBGP correctness
   and safe operation in hop-by-hop routed networks.  When routers
   within an AS apply different criteria in selecting routes, they can
   arrive at inconsistent route selections.  This can lead to the
   formation of forwarding loops unless some form of tunneled forwarding
   is used to prevent "core" routers from making a (potentially
   inconsistent) forwarding decision based on the IP header.

   This specification uses the state of a peering session as an input to
   the selection criteria, depreferencing routes that are associated
   with a session that has gone down but that have not yet aged out.
   Since different routers within an AS might have different notions as
   to whether their respective sessions with a given peer are up or
   down, they might apply different selection criteria to routes from
   that peer.  This could result in a forwarding loop forming between
   such routers.

   For an example of such a forwarding loop, consider the following
   simple topology:


   A ---- B ---- C ------------------------- D
   ^                                         ^
   |                                         |
   R1                                        R2

                                  Figure 1

   In this example, A - D are routers with a full mesh of IBGP sessions
   between them (the sessions are not shown).  The short links have unit
   cost, the long link has cost 5.  Routers A and D are AS border
   routers, each advertising some route, R, with the same LOCAL_PREF
   into the AS: denoted R1 and R2 in the diagram.  In ordinary
   operation, it can be seen that routers B and C will select R1 for
   forwarding and will forward toward A.

   Suppose that the session between A and B goes down for some reason,
   and it stays down long enough for LLGR processing to be invoked on B.
   Then, on B, route R1 will be depreferenced, leading to the selection
   of R2 by B.  However, C will continue to prefer R1.  In this case, it
   can be seen that a forwarding loop for packets destined to R would
   form between B and C.  (We note that other forwarding loop scenarios
   can be constructed for conventional GR, but these are generally
   considered less severe since GR can remain in effect for a much more
   limited interval.)

   The potential benefits of this specification can outweigh the risks
   discussed above, as long as care is exercised in deployment.  The
   cardinal rule to be followed is that if a given set of routes is
   being used within an AS for hop-by-hop forwarding, enabling LLGR
   procedures is not recommended.  If tunneled forwarding (such as MPLS)
   is used within the AS, or if routes are being used for purposes other
   than hop-by-hop forwarding, less caution is needed; however, the
   operator should still carefully consider the consequences of enabling
   LLGR.

6.  Security Considerations

   The security implications of the LLGR mechanism defined in this
   document are akin to those incurred by the maintenance of stale
   routing information within a network.  However, since the retention
   time may be much longer, the window during which certain attacks are
   feasible may substantially increase.  This is particularly relevant
   when considering the maintenance of routing information that is used
   for service segregation, such as MPLS label entries.

   For MPLS VPN services, the effectiveness of the traffic isolation
   between VPNs relies on the correctness of the MPLS labels between
   ingress and egress PEs.  In particular, when an egress PE withdraws a
   label L1 allocated to a VPN1 route, this label must not be assigned
   to a VPN route of a different VPN until all ingress PEs stop using
   the old VPN1 route using L1.

   Such a corner case may happen today if the propagation of VPN routes
   by BGP messages between PEs takes more time than the label
   reallocation delay on a PE.  Given that we can generally bound the
   worst-case BGP propagation time to a few minutes (for example, 2-5
   minutes), the security breach will not occur if PEs are designed to
   not reallocate a previously used and withdrawn label before a few
   minutes.

   The problem is made worse with BGP GR between PEs because VPN routes
   can be stalled for a longer period of time (for example, 20 minutes).

   This is further aggravated by the LLGR extension specified in this
   document because VPN routes can be stalled for a much longer period
   of time (for example, 2 hours, 1 day).

   In order to exploit the vulnerability described above, an attacker
   needs to engineer a specific LLGR state between two PE devices and
   also cause the label reallocation to occur such that the two
   topologies overlap.  To avoid the potential for a VPN breach, the
   operator should ensure that the lower bound for label reuse is
   greater than the upper bound on the LLST before enabling LLGR for a
   VPN address family.  Section 4.2 discusses the provision of an upper
   bound on LLST.  Details of features for setting a lower bound on
   label reuse time are beyond the scope of this document; however,
   factors that might need to be taken into account when setting this
   value include:

   *  The load of the BGP route churn on a PE (in terms of the number of
      VPN labels advertised and the churn rate).

   *  The label allocation policy on the PE, which possibly depends upon
      the size of the pool of the VPN labels (which can be restricted by
      hardware considerations or other MPLS usages), the label
      allocation scheme (for example, per route or per VRF/CE), and the
      reallocation policy (for example, least recently used label).

   Note that [RFC4781], which defines the Graceful Restart Mechanism for
   BGP with MPLS, is also applicable to LLGR.

7.  Examples of Operation

   For illustrative purposes, we present a few examples of how this
   specification might be used in practice.  These examples are neither
   exhaustive nor normative.

   Consider the following scenario: A border router, ASBR1, has an IBGP
   peering with a route reflector, RR1, from which it learns routes.  It
   has an EBGP peering with an external peer, EXT, to which it
   advertises those routes.  The external peer has advertised the GR and
   LLGR Capabilities to ASBR1.  ASBR1 is configured to support GR and
   LLGR on its sessions with RR1 and EXT.  RR1 advertises a GR Restart
   Time of 1 (second) and an LLST of 3600 (seconds):

    +==========+=====================================================+
    | Time     | Event                                               |
    +==========+=====================================================+
    | t        | ASBR1's IBGP session with RR fails.  ASBR1 retains  |
    |          | RR's routes according to the rules of GR [RFC4724]. |
    +----------+-----------------------------------------------------+
    | t+1      | GR Restart Time expires.  ASBR1 transitions RR's    |
    |          | routes to long-lived stale routes by attaching the  |
    |          | LLGR_STALE community and depreferencing them.       |
    |          | However, since it has no backup routes, it          |
    |          | continues to make use of them.  It re-announces     |
    |          | them to EXT with the LLGR_STALE community attached. |
    +----------+-----------------------------------------------------+
    | t+1+3600 | LLST expires.  ASBR1 removes RR's stale routes from |
    |          | its own RIB and sends BGP updates to withdraw them  |
    |          | from EXT.                                           |
    +----------+-----------------------------------------------------+

                                 Table 1

   Next, imagine the same scenario, but suppose RR1 advertised a GR
   Restart Time of zero, effectively disabling GR.  Equally, ASBR1 could
   have used a local configuration to override RR1's offered Restart
   Time, setting it to a locally configured value of zero:

   +==========+=======================================================+
   | Time     | Event                                                 |
   +==========+=======================================================+
   | t        | ASBR1's IBGP session with RR fails.  ASBR1            |
   |          | transitions RR's routes to long-lived stale routes by |
   |          | attaching the LLGR_STALE community and depreferencing |
   |          | them.  However, since it has no backup routes, it     |
   |          | continues to make use of them.  It re-announces them  |
   |          | to EXT with the LLGR_STALE community attached.        |
   +----------+-------------------------------------------------------+
   | t+0+3600 | LLST expires.  ASBR1 removes RR's stale routes from   |
   |          | its own RIB and sends BGP updates to withdraw them    |
   |          | from EXT.                                             |
   +----------+-------------------------------------------------------+

                                 Table 2

   Next, imagine the original scenario, but consider that the ASBR1-RR1
   session comes back up and becomes synchronized 180 seconds after the
   failure was detected:

     +=========+=====================================================+
     | Time    | Event                                               |
     +=========+=====================================================+
     | t       | ASBR1's IBGP session with RR fails.  ASBR1 retains  |
     |         | RR's routes according to the rules of GR [RFC4724]. |
     +---------+-----------------------------------------------------+
     | t+1     | GR Restart Time expires.  ASBR1 transitions RR's    |
     |         | routes to long-lived stale routes by attaching the  |
     |         | LLGR_STALE community and depreferencing them.       |
     |         | However, since it has no backup routes, it          |
     |         | continues to make use of them.  It re-announces     |
     |         | them to EXT with the LLGR_STALE community attached. |
     +---------+-----------------------------------------------------+
     | t+1+179 | Session is re-established and resynchronized.       |
     |         | ASBR1 removes the LLGR_STALE community from RR1's   |
     |         | routes and re-announces them to EXT with the        |
     |         | LLGR_STALE community removed.                       |
     +---------+-----------------------------------------------------+

                                  Table 3

   Finally, imagine the original scenario, but consider that EXT has not
   advertised the LLGR Capability to ASBR1:

    +==========+======================================================+
    | Time     | Event                                                |
    +==========+======================================================+
    | t        | ASBR1's IBGP session with RR fails.  ASBR1 retains   |
    |          | RR's routes according to the rules of GR [RFC4724].  |
    +----------+------------------------------------------------------+
    | t+1      | GR Restart Time expires.  ASBR1 transitions RR's     |
    |          | routes to long-lived stale routes by attaching the   |
    |          | LLGR_STALE community and depreferencing them.        |
    |          | However, since it has no backup routes, it continues |
    |          | to make use of them.  It withdraws them from EXT.    |
    +----------+------------------------------------------------------+
    | t+1+3600 | LLST expires.  ASBR1 removes RR's stale routes from  |
    |          | its own RIB.                                         |
    +----------+------------------------------------------------------+

                                  Table 4

8.  IANA Considerations

   This document defines a BGP capability called the "Long-Lived
   Graceful Restart Capability".  IANA has assigned a value of 71 from
   the "Capability Codes" registry.

   This document introduces two BGP well-known communities:

   *  the first called "LLGR_STALE" for marking long-lived stale routes,
      and

   *  the second called "NO_LLGR" for marking routes that should not be
      retained if stale.

   IANA has assigned these well-known community values 0xFFFF0006 and
   0xFFFF0007, respectively, from the "BGP Well-known Communities"
   registry.

   IANA has established a registry called the "Long-Lived Graceful
   Restart Flags for Address Family" registry under the "Border Gateway
   Protocol (BGP) Parameters" group.  The registration procedures are
   Standards Action (see [RFC8126]).  The registry is initially
   populated as follows:

     +==============+=======================+============+===========+
     | Bit Position | Name                  | Short Name | Reference |
     +==============+=======================+============+===========+
     | 0            | Preservation of state | F          | RFC 9494  |
     +--------------+-----------------------+------------+-----------+
     | 1-7          | Unassigned            |            |           |
     +--------------+-----------------------+------------+-----------+

                                  Table 5

9.  References

9.1.  Normative References

   [RFC1997]  Chandra, R., Traina, P., and T. Li, "BGP Communities
              Attribute", RFC 1997, DOI 10.17487/RFC1997, August 1996,
              <https://www.rfc-editor.org/info/rfc1997>.

   [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119,
              DOI 10.17487/RFC2119, March 1997,
              <https://www.rfc-editor.org/info/rfc2119>.

   [RFC4271]  Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A
              Border Gateway Protocol 4 (BGP-4)", RFC 4271,
              DOI 10.17487/RFC4271, January 2006,
              <https://www.rfc-editor.org/info/rfc4271>.

   [RFC4724]  Sangli, S., Chen, E., Fernando, R., Scudder, J., and Y.
              Rekhter, "Graceful Restart Mechanism for BGP", RFC 4724,
              DOI 10.17487/RFC4724, January 2007,
              <https://www.rfc-editor.org/info/rfc4724>.

   [RFC4760]  Bates, T., Chandra, R., Katz, D., and Y. Rekhter,
              "Multiprotocol Extensions for BGP-4", RFC 4760,
              DOI 10.17487/RFC4760, January 2007,
              <https://www.rfc-editor.org/info/rfc4760>.

   [RFC5492]  Scudder, J. and R. Chandra, "Capabilities Advertisement
              with BGP-4", RFC 5492, DOI 10.17487/RFC5492, February
              2009, <https://www.rfc-editor.org/info/rfc5492>.

   [RFC6368]  Marques, P., Raszuk, R., Patel, K., Kumaki, K., and T.
              Yamagata, "Internal BGP as the Provider/Customer Edge
              Protocol for BGP/MPLS IP Virtual Private Networks (VPNs)",
              RFC 6368, DOI 10.17487/RFC6368, September 2011,
              <https://www.rfc-editor.org/info/rfc6368>.

   [RFC8174]  Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC
              2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
              May 2017, <https://www.rfc-editor.org/info/rfc8174>.

   [RFC8538]  Patel, K., Fernando, R., Scudder, J., and J. Haas,
              "Notification Message Support for BGP Graceful Restart",
              RFC 8538, DOI 10.17487/RFC8538, March 2019,
              <https://www.rfc-editor.org/info/rfc8538>.

9.2.  Informative References

   [RFC4364]  Rosen, E. and Y. Rekhter, "BGP/MPLS IP Virtual Private
              Networks (VPNs)", RFC 4364, DOI 10.17487/RFC4364, February
              2006, <https://www.rfc-editor.org/info/rfc4364>.

   [RFC4761]  Kompella, K., Ed. and Y. Rekhter, Ed., "Virtual Private
              LAN Service (VPLS) Using BGP for Auto-Discovery and
              Signaling", RFC 4761, DOI 10.17487/RFC4761, January 2007,
              <https://www.rfc-editor.org/info/rfc4761>.

   [RFC4781]  Rekhter, Y. and R. Aggarwal, "Graceful Restart Mechanism
              for BGP with MPLS", RFC 4781, DOI 10.17487/RFC4781,
              January 2007, <https://www.rfc-editor.org/info/rfc4781>.

   [RFC5880]  Katz, D. and D. Ward, "Bidirectional Forwarding Detection
              (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010,
              <https://www.rfc-editor.org/info/rfc5880>.

   [RFC8126]  Cotton, M., Leiba, B., and T. Narten, "Guidelines for
              Writing an IANA Considerations Section in RFCs", BCP 26,
              RFC 8126, DOI 10.17487/RFC8126, June 2017,
              <https://www.rfc-editor.org/info/rfc8126>.

   [RFC8955]  Loibl, C., Hares, S., Raszuk, R., McPherson, D., and M.
              Bacher, "Dissemination of Flow Specification Rules",
              RFC 8955, DOI 10.17487/RFC8955, December 2020,
              <https://www.rfc-editor.org/info/rfc8955>.

Acknowledgements

   We would like to thank Nabil Bitar, Martin Djernaes, Roberto
   Fragassi, Jeffrey Haas, Jakob Heitz, Daniam Henriques, Nicolai
   Leymann, Mike McBride, Paul Mattes, John Medamana, Pranav Mehta, Han
   Nguyen, Saikat Ray, Valery Smyslov, and Bo Wu for their valuable
   input and contributions to the discussion and solution.

Contributors

   Clarence Filsfils
   Cisco Systems
   1150 Brussels
   Belgium
   Email: cf@cisco.com


   Pradosh Mohapatra
   Sproute Networks
   Email: mpradosh@yahoo.com


   Yakov Rekhter


   Eric Rosen
   Email: erosen52@gmail.com


   Rob Shakir
   Google, Inc.
   1600 Amphitheatre Parkway
   Mountain View, CA 94043
   United States of America
   Email: robjs@google.com


   Adam Simpson
   Nokia
   Email: adam.1.simpson@nokia.com


Authors' Addresses

   James Uttaro
   Independent Contributor
   Email: juttaro@ieee.org


   Enke Chen
   Palo Alto Networks
   Email: enchen@paloaltonetworks.com


   Bruno Decraene
   Orange
   Email: bruno.decraene@orange.com


   John G. Scudder
   Juniper Networks
   Email: jgs@juniper.net
  1. RFC 9494