RFC 8706




Internet Engineering Task Force (IETF)                       L. Ginsberg
Request for Comments: 8706                                      P. Wells
Obsoletes: 5306                                      Cisco Systems, Inc.
Category: Standards Track                                  February 2020
ISSN: 2070-1721


                      Restart Signaling for IS-IS

Abstract



   This document describes a mechanism for a restarting router to signal
   to its neighbors that it is restarting, allowing them to reestablish
   their adjacencies without cycling through the DOWN state while still
   correctly initiating database synchronization.

   This document additionally describes a mechanism for a router to
   signal its neighbors that it is preparing to initiate a restart while
   maintaining forwarding-plane state.  This allows the neighbors to
   maintain their adjacencies until the router has restarted but also
   allows the neighbors to bring the adjacencies down in the event of
   other topology changes.

   This document additionally describes a mechanism for a restarting
   router to determine when it has achieved Link State Protocol Data
   Unit (LSP) database synchronization with its neighbors and a
   mechanism to optimize LSP database synchronization while minimizing
   transient routing disruption when a router starts.

   This document obsoletes RFC 5306.

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/rfc8706.

Copyright Notice



   Copyright (c) 2020 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 Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

Table of Contents



   1.  Overview
   2.  Conventions Used in This Document
     2.1.  Requirements Language
   3.  Approach
     3.1.  Timers
     3.2.  Restart TLV
       3.2.1.  Use of RR and RA Bits
       3.2.2.  Use of the SA Bit
       3.2.3.  Use of PR and PA Bits
     3.3.  Adjacency (Re)Acquisition
       3.3.1.  Adjacency Reacquisition during Restart
       3.3.2.  Adjacency Acquisition during Start
       3.3.3.  Multiple Levels
     3.4.  Database Synchronization
       3.4.1.  LSP Generation and Flooding and SPF Computation
   4.  State Tables
     4.1.  Running Router
     4.2.  Restarting Router
     4.3.  Starting Router
   5.  IANA Considerations
   6.  Security Considerations
   7.  Manageability Considerations
   8.  Normative References
   Appendix A.  Summary of Changes from RFC 5306
   Acknowledgements

   Authors' Addresses



1.  Overview



   The Intermediate System to Intermediate System (IS-IS) routing
   protocol [RFC1195] [ISO10589] is a link state intra-domain routing
   protocol.  Normally, when an IS-IS router is restarted, temporary
   disruption of routing occurs due to events in both the restarting
   router and the neighbors of the restarting router.

   The router that has been restarted computes its own routes before
   achieving database synchronization with its neighbors.  The results
   of this computation are likely to be non-convergent with the routes
   computed by other routers in the area/domain.

   Neighbors of the restarting router detect the restart event and cycle
   their adjacencies with the restarting router through the DOWN state.
   The cycling of the adjacency state causes the neighbors to regenerate
   their LSPs describing the adjacency concerned.  This in turn causes a
   temporary disruption of routes passing through the restarting router.

   In certain scenarios, the temporary disruption of the routes is
   highly undesirable.  This document describes mechanisms to avoid or
   minimize the disruption due to both of these causes.

   When an adjacency is reinitialized as a result of a neighbor
   restarting, a router does three things:

   1.  It causes its own LSP(s) to be regenerated, thus triggering
       Shortest Path First (SPF) runs throughout the area (or in the
       case of Level 2, throughout the domain).

   2.  It sets SRMflags on its own LSP database on the adjacency
       concerned.



   3.  In the case of a Point-to-Point link, it transmits a complete set
       of Complete Sequence Number PDUs (CSNPs), over the adjacency.



   In the case of a restarting router process, the first of these is
   highly undesirable, but the second is essential in order to ensure
   synchronization of the LSP database.

   The third action above minimizes the number of LSPs that must be
   exchanged and, if made reliable, provides a means of determining when
   the LSP databases of the neighboring routers have been synchronized.
   This is desirable whether or not the router is being restarted (so
   that the overload bit can be cleared in the router's own LSP, for
   example).

   This document describes a mechanism for a restarting router to signal
   to its neighbors that it is restarting.  The mechanism further allows
   the neighbors to reestablish their adjacencies with the restarting
   router without cycling through the DOWN state while still correctly
   initiating database synchronization.

   This document additionally describes a mechanism for a restarting
   router to determine when it has achieved LSP database synchronization
   with its neighbors and a mechanism to optimize LSP database
   synchronization and minimize transient routing disruption when a
   router starts.

   It is assumed that the three-way handshake [RFC5303] is being used on
   Point-to-Point circuits.

2.  Conventions Used in This Document

   If the control and forwarding functions in a router can be maintained
   independently, it is possible for the forwarding function state to be
   maintained across a resumption of control function operations.  This
   functionality is assumed when the terms "restart/restarting" are used
   in this document.

   The terms "start/starting" are used to refer to a router in which the
   control function has either commenced operations for the first time
   or has resumed operations, but the forwarding functions have not been
   maintained in a prior state.

   The terms "(re)start/(re)starting" are used when the text is
   applicable to both a "starting" and a "restarting" router.

   The terms "normal IIH" or "IIH normal" refer to IS-IS Hellos (IIHs)
   in which the Restart TLV (defined later in this document) has no
   flags set.

2.1.  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.  Approach

3.1.  Timers



   Three additional timers (T1, T2, and T3) are required to support the
   mechanisms defined in this document.  Timers T1 and T2 are used both
   by a restarting router and a starting router.  Timer T3 is used only
   by a restarting router.

   NOTE: These timers are NOT applicable to a router that is preparing
   to do a planned restart.

   An instance of the timer T1 is maintained per interface and indicates
   the time after which an unacknowledged (re)start attempt will be
   repeated.  A typical value is 3 seconds.

   An instance of the timer T2 is maintained for each LSP database
   (LSPDB) present in the system.  For example, for a Level 1/2 system,
   there will be an instance of the timer T2 for Level 1 and an instance
   for Level 2.  This is the maximum time that the system will wait for
   LSPDB synchronization.  A typical value is 60 seconds.

   A single instance of the timer T3 is maintained for the entire
   system.  It indicates the time after which the router will declare
   that it has failed to achieve database synchronization (by setting
   the overload bit in its own LSP).  This is initialized to 65535
   seconds but is set to the minimum of the remaining times of received
   IIHs containing a Restart TLV with the Restart Acknowledgement (RA)
   set and an indication that the neighbor has an adjacency in the UP
   state to the restarting router.  (See item a in Section 3.2.1.)

3.2.  Restart TLV



   A new TLV is defined to be included in IIH PDUs.  The TLV includes
   flags that are used to convey information during a (re)start.  The
   absence of this TLV indicates that the sender supports none of the
   functionality defined in this document.  Therefore, if a router
   supports any of the functionality defined in this document it MUST
   include this TLV in all transmitted IIHs.

   Type:
      211

   Length:
      Number of octets in the Value field (1 to (3 + ID Length))

   Value:
                                            No. of octets
              +-----------------------+
              |   Flags               |     1
              +-----------------------+
              | Remaining Time        |     2
              +-----------------------+
              | Restarting Neighbor ID|     ID Length
              +-----------------------+

      Flags (1 octet)
               0  1  2  3  4  5  6  7
              +--+--+--+--+--+--+--+--+
              |Reserved|PA|PR|SA|RA|RR|
              +--+--+--+--+--+--+--+--+

         RR -  Restart Request
         RA -  Restart Acknowledgement
         SA -  Suppress adjacency advertisement
         PR -  Restart is planned
         PA -  Planned restart acknowledgement

      Remaining Time (2 octets)
         Remaining Holding Time (in seconds).

         Required when the RA, PR, or PA bit is set.  Otherwise, this
         field SHOULD be omitted when sent and MUST be ignored when
         received.

      Restarting Neighbor System ID (ID Length octets)
         The System ID of the neighbor to which an RA/PA refers.

         Required when the RA or PA bit is set.  Otherwise, this field
         SHOULD be omitted when sent and MUST be ignored when received.

         Note: Very early draft versions of the restart functionality
         did not include the Restarting Neighbor System ID in the TLV.
         RFC 5306 allowed for the possibility of interoperating with
         legacy implementations by stating that a router that is
         expecting an RA on a LAN circuit should assume that the
         acknowledgement is directed at the local system if the TLV is
         received with RA set and Restarting Neighbor System ID is not
         present.  It is an implementation choice whether to continue to
         accept (on a LAN) a TLV with RA set and Restarting Neighbor
         System ID absent.  Note that the omission of the Restarting
         Neighbor System ID only introduces ambiguity in the case where
         there are multiple systems on a LAN simultaneously performing
         restart.

   The RR and SA flags may both be set in the TLV under the conditions
   described in Section 3.3.2.  All other combinations where multiple
   flags are set are invalid and MUST NOT be transmitted.  Received TLVs
   that have invalid flag combinations set MUST be ignored.

3.2.1.  Use of RR and RA Bits



   The RR bit is used by a (re)starting router to signal to its
   neighbors that a (re)start is in progress, that an existing adjacency
   SHOULD be maintained even under circumstances when the normal
   operation of the adjacency state machine would require the adjacency
   to be reinitialized, to request a set of CSNPs, and to request
   setting of the SRMflags.

   The RA bit is sent by the neighbor of a (re)starting router to
   acknowledge the receipt of a Restart TLV with the RR bit set.

   When the neighbor of a (re)starting router receives an IIH with the
   Restart TLV having the RR bit set, if there exists on this interface
   an adjacency in the UP state with the same System ID and, in the case
   of a LAN circuit, with the same source LAN address, then irrespective
   of the other contents of the "Intermediate System Neighbors" option
   (LAN circuits) or the "Point-to-Point Three-Way Adjacency" option
   (Point-to-Point circuits):

   a.  the state of the adjacency is not changed.  If this is the first
       IIH with the RR bit set that this system has received associated
       with this adjacency, then the adjacency is marked as being in
       "Restart mode" and the adjacency Holding Time is refreshed --
       otherwise, the Holding Time is not refreshed.  The Remaining Time
       transmitted according to (b) below MUST reflect the actual time
       after which the adjacency will now expire.  Receipt of an IIH
       with the RR bit reset will clear the "Restart mode" state.  This
       procedure allows the restarting router to cause the neighbor to
       maintain the adjacency long enough for restart to successfully
       complete while also preventing repetitive restarts from
       maintaining an adjacency indefinitely.  Whether or not an
       adjacency is marked as being in "Restart mode" has no effect on
       adjacency state transitions.

   b.  immediately (i.e., without waiting for any currently running
       timer interval to expire but with a small random delay of a few
       tens of milliseconds on LANs to avoid "storms") transmit over the
       corresponding interface an IIH including the Restart TLV with the
       RR bit clear and the RA bit set, in the case of Point-to-Point
       adjacencies having updated the "Point-to-Point Three-Way
       Adjacency" option to reflect any new values received from the
       (re)starting router.  (This allows a restarting router to quickly
       acquire the correct information to place in its hellos.)  The
       Remaining Time MUST be set to the current time (in seconds)
       before the holding timer on this adjacency is due to expire.  If
       the corresponding interface is a LAN interface, then the
       Restarting Neighbor System ID SHOULD be set to the System ID of
       the router from which the IIH with the RR bit set was received.
       This is required to correctly associate the acknowledgement and
       Holding Time in the case where multiple systems on a LAN restart
       at approximately the same time.  This IIH SHOULD be transmitted
       before any LSPs or SNPs are transmitted as a result of the
       receipt of the original IIH.

   c.  if the corresponding interface is a Point-to-Point interface, or
       if the receiving router has the highest LnRouterPriority (with
       the highest source Media Access Control (MAC) address breaking
       ties) among those routers to which the receiving router has an
       adjacency in the UP state on this interface whose IIHs contain
       the Restart TLV, excluding adjacencies to all routers that are
       considered in "Restart mode" (note the actual Designated
       Intermediate System (DIS) is NOT changed by this process),
       initiate the transmission over the corresponding interface of a
       complete set of CSNPs, and set SRMflags on the corresponding
       interface for all LSPs in the local LSP database.

   Otherwise (i.e., if there was no adjacency in the UP state to the
   System ID in question), process the IIH as normal by reinitializing
   the adjacency and setting the RA bit in the returned IIH.

3.2.2.  Use of the SA Bit



   The SA bit is used by a starting router to request that its neighbor
   suppress advertisement of the adjacency to the starting router in the
   neighbor's LSPs.

   A router that is starting has no maintained forwarding function
   state.  This may or may not be the first time the router has started.
   If this is not the first time the router has started, copies of LSPs
   generated by this router in its previous incarnation may exist in the
   LSP databases of other routers in the network.  These copies are
   likely to appear "newer" than LSPs initially generated by the
   starting router due to the reinitialization of LSP fragment sequence
   numbers by the starting router.  This may cause temporary blackholes
   to occur until the normal operation of the update process causes the
   starting router to regenerate and flood copies of its own LSPs with
   higher sequence numbers.  The temporary blackholes can be avoided if
   the starting router's neighbors suppress advertising an adjacency to
   the starting router until the starting router has been able to
   propagate newer versions of LSPs generated by previous incarnations.

   When a router receives an IIH with the Restart TLV having the SA bit
   set, if there exists on this interface an adjacency in the UP state
   with the same System ID and, in the case of a LAN circuit, with the
   same source LAN address, then the router MUST suppress advertisement
   of the adjacency to the neighbor in its own LSPs.  Until an IIH with
   the SA bit clear has been received, the neighbor advertisement MUST
   continue to be suppressed.  If the adjacency transitions to the UP
   state, the new adjacency MUST NOT be advertised until an IIH with the
   SA bit clear has been received.

   Note that a router that suppresses advertisement of an adjacency MUST
   NOT
use this adjacency when performing its SPF calculation.  In
   particular, if an implementation follows the example guidelines
   presented in [ISO10589], Annex C.2.5, Step 0:b) "pre-load TENT with
   the local adjacency database", the suppressed adjacency MUST NOT be
   loaded into TENT.

3.2.3.  Use of PR and PA Bits



   The PR bit is used by a router that is planning to initiate a restart
   to signal to its neighbors that it will be restarting.  The router
   sending an IIH with PR bit set SHOULD set the Remaining Time to a
   value greater than the expected control-plane restart time.  The PR
   bit SHOULD remain set in IIHs until the restart is initiated.

   The PA bit is sent by the neighbor of a router planning to restart to
   acknowledge receipt of a Restart TLV with the PR bit set.

   When the neighbor of a router planning a restart receives an IIH with
   the Restart TLV having the PR bit set, if there exists on this
   interface an adjacency in the UP state with the same System ID and,
   in the case of a LAN circuit, with the same source LAN address, then:

   a.  if this is the first IIH with the PR bit set that this system has
       received associated with this adjacency, then the adjacency is
       marked as being in Planned Restart State and the adjacency
       Holding Time is refreshed -- otherwise, the Holding Time is not
       refreshed.  The Holding Time SHOULD be set to the Remaining Time
       specified in the received IIH with PR set.  The Remaining Time
       transmitted according to (b) below MUST reflect the actual time
       after which the adjacency will now expire.  Receipt of an IIH
       with the PR bit reset will clear the Planned Restart State and
       cause the receiving router to set the adjacency Holding Time to
       the locally configured value.  This procedure allows the router
       planning a restart to cause the neighbor to maintain the
       adjacency long enough for restart to successfully complete.
       Whether or not an adjacency is marked as being in Planned Restart
       State has no effect on adjacency state transitions.

   b.  immediately (i.e., without waiting for any currently running
       timer interval to expire, but with a small random delay of a few
       tens of milliseconds on LANs to avoid "storms") transmit over the
       corresponding interface an IIH including the Restart TLV with the
       PR bit clear and the PA bit set.  The Remaining Time MUST be set
       to the current time (in seconds) before the holding timer on this
       adjacency is due to expire.  If the corresponding interface is a
       LAN interface, then the Restarting Neighbor System ID SHOULD be
       set to the System ID of the router from which the IIH with the PR
       bit set was received.  This is required to correctly associate
       the acknowledgement and Holding Time in the case where multiple
       systems on a LAN are planning a restart at approximately the same
       time.

   NOTE: Receipt of an IIH with PA bit set indicates to the router
   planning a restart that the neighbor is aware of the planned restart
   and -- in the absence of topology changes as described below -- will
   maintain the adjacency for the Remaining Time included in the IIH
   with PA set.

   By definition, a restarting router maintains forwarding state across
   the control-plane restart (see Section 2).  But while a control-plane
   restart is in progress, it is expected that the restarting router
   will be unable to respond to topology changes.  It is therefore
   useful to signal a planned restart so that the neighbors of the
   restarting router can determine whether it is safe to maintain the
   adjacency if other topology changes occur prior to the completion of
   the restart.  Signaling a planned restart in the absence of
   maintained forwarding-plane state is likely to lead to significant
   traffic loss and MUST NOT be done.

   Neighbors of the router that have signaled planned restart SHOULD
   maintain the adjacency in a Planned Restart State until it receives
   an IIH with the RR bit set, it receives an IIH with both PR and RR
   bits clear, or the adjacency Holding Time expires -- whichever occurs
   first.  Neighbors that choose not to follow the recommended behavior
   need to consider the impact on traffic delivery of not using the
   restarting router for forwarding traffic during the restart period.

   While the adjacency is in Planned Restart State, some or all of the
   following actions MAY be taken:

   a.  If additional topology changes occur, the adjacency that is in
       Planned Restart State MAY be brought down even though the Holding
       Time has not yet expired.  Given that the neighbor that has
       signaled a planned restart is not expected to update its
       forwarding plane in response to signaling of the topology changes
       (since it is restarting) traffic that transits that node is at
       risk of being improperly forwarded.  On a LAN circuit, if the
       router in Planned Restart State is the DIS at any supported
       level, the adjacency or adjacencies SHOULD be brought down
       whenever any LSP update is either generated or received so as to
       trigger a new DIS election.  Failure to do so will compromise the
       reliability of the update process on that circuit.  What other
       criteria are used to determine what topology changes will trigger
       bringing the adjacency down is a local implementation decision.

   b.  If a Bidirectional Forwarding Detection (BFD) [RFC5880] Session
       to the neighbor that signals a planned restart is in the UP state
       and subsequently goes down, the event MAY be ignored since it is
       possible this is an expected side effect of the restart.  Use of
       the Control-Plane Independent state as signaled in BFD control
       packets SHOULD be considered in the decision to ignore a BFD
       Session DOWN event.

   c.  On a Point-to-Point circuit, transmission of LSPs, CSNPs, and
       Partial Sequence Number PDU (PSNPs) MAY be suppressed.  It is
       expected that the PDUs will not be received.

   Use of the PR bit provides a means to safely support restart periods
   that are significantly longer than standard Holding Times.

3.3.  Adjacency (Re)Acquisition



   Adjacency (re)acquisition is the first step in (re)initialization.
   Restarting and starting routers will make use of the RR bit in the
   Restart TLV, though each will use it at different stages of the
   (re)start procedure.

3.3.1.  Adjacency Reacquisition during Restart



   The restarting router explicitly notifies its neighbor that the
   adjacency is being reacquired and, hence, that it SHOULD NOT
   reinitialize the adjacency.  This is achieved by setting the RR bit
   in the Restart TLV.  When the neighbor of a restarting router
   receives an IIH with the Restart TLV having the RR bit set, if there
   exists on this interface an adjacency in the UP state with the same
   System ID and, in the case of a LAN circuit, with the same source LAN
   address, then the procedures described in Section 3.2.1 are followed.

   A router that does not support the restart capability will ignore the
   Restart TLV and reinitialize the adjacency as normal, returning an
   IIH without the Restart TLV.

   On restarting, a router initializes the timer T3, starts the timer T2
   for each LSPDB, and for each interface (and in the case of a LAN
   circuit, for each level) starts the timer T1 and transmits an IIH
   containing the Restart TLV with the RR bit set.

   On a Point-to-Point circuit, the restarting router SHOULD set the
   "Adjacency Three-Way State" to "Init", because the receipt of the
   acknowledging IIH (with RA set) MUST cause the adjacency to enter the
   UP state immediately.

   On a LAN circuit, the LAN-ID assigned to the circuit SHOULD be the
   same as that used prior to the restart.  In particular, for any
   circuits for which the restarting router was previously DIS, the use
   of a different LAN-ID would necessitate the generation of a new set
   of pseudonode LSPs and corresponding changes in all the LSPs
   referencing them from other routers on the LAN.  By preserving the
   LAN-ID across the restart, this churn can be prevented.  To enable a
   restarting router to learn the LAN-ID used prior to restart, the LAN-
   ID specified in an IIH with RR set MUST be ignored.

   Transmission of "normal IIHs" is inhibited until the conditions
   described below are met (in order to avoid causing an unnecessary
   adjacency initialization).  Upon expiry of the timer T1, it is
   restarted and the IIH is retransmitted as above.

   When a restarting router receives an IIH a local adjacency is
   established as usual, and if the IIH contains a Restart TLV with the
   RA bit set (and on LAN circuits with a Restart Neighbor System ID
   that matches that of the local system), the receipt of the
   acknowledgement over that interface is noted.  When the RA bit is set
   and the state of the remote adjacency is UP, then the timer T3 is set
   to the minimum of its current value and the value of the Remaining
   Time field in the received IIH.

   On a Point-to-Point link, receipt of an IIH not containing the
   Restart TLV is also treated as an acknowledgement, since it indicates
   that the neighbor is not restart capable.  However, since no CSNP is
   guaranteed to be received over this interface, the timer T1 is
   canceled immediately without waiting for a complete set of CSNPs.
   Synchronization may therefore be deemed complete even though there
   are some LSPs that are held (only) by this neighbor (see
   Section 3.4).  In this case, we also want to be certain that the
   neighbor will reinitialize the adjacency in order to guarantee that
   the SRMflags have been set on its database, thus ensuring eventual
   LSPDB synchronization.  This is guaranteed to happen except in the
   case where the Adjacency Three-Way State in the received IIH is UP
   and the Neighbor Extended Local Circuit ID matches the Extended Local
   Circuit ID assigned by the restarting router.  In this case, the
   restarting router MUST force the adjacency to reinitialize by setting
   the local Adjacency Three-Way State to DOWN and sending a normal IIH.

   In the case of a LAN interface, receipt of an IIH not containing the
   Restart TLV is unremarkable since synchronization can still occur so
   long as at least one of the non-restarting neighboring routers on the
   LAN supports restart.  Therefore, T1 continues to run in this case.
   If none of the neighbors on the LAN are restart capable, T1 will
   eventually expire after the locally defined number of retries.

   In the case of a Point-to-Point circuit, the LocalCircuitID and
   Extended Local Circuit ID information contained in the IIH can be
   used immediately to generate an IIH containing the correct three-way
   handshake information.  The presence of Neighbor Extended Local
   Circuit ID information that does not match the value currently in use
   by the local system is ignored (since the IIH may have been
   transmitted before the neighbor had received the new value from the
   restarting router), but the adjacency remains in the initializing
   state until the correct information is received.

   In the case of a LAN circuit, the source neighbor information (e.g.,
   SNPAAddress) is recorded and used for adjacency establishment and
   maintenance as normal.

   When BOTH a complete set of CSNPs (for each active level, in the case
   of a Point-to-Point circuit) and an acknowledgement have been
   received over the interface, the timer T1 is canceled.

   Once the timer T1 has been canceled, subsequent IIHs are transmitted
   according to the normal algorithms but including the Restart TLV with
   both RR and RA clear.

   If a LAN contains a mixture of systems, only some of which support
   the new algorithm, database synchronization is still guaranteed, but
   the "old" systems will have reinitialized their adjacencies.

   If an interface is active but does not have any neighboring router
   reachable over that interface, the timer T1 would never be canceled,
   and according to Section 3.4.1.1, the SPF would never be run.
   Therefore, timer T1 is canceled after some predetermined number of
   expirations (which MAY be 1).

3.3.2.  Adjacency Acquisition during Start



   The starting router wants to ensure that in the event that a
   neighboring router has an adjacency to the starting router in the UP
   state (from a previous incarnation of the starting router), this
   adjacency is reinitialized.  The starting router also wants
   neighboring routers to suppress advertisement of an adjacency to the
   starting router until LSP database synchronization is achieved.  This
   is achieved by sending IIHs with the RR bit clear and the SA bit set
   in the Restart TLV.  The RR bit remains clear and the SA bit remains
   set in subsequent transmissions of IIHs until the adjacency has
   reached the UP state and the initial T1 timer interval (see below)
   has expired.

   Receipt of an IIH with the RR bit clear will result in the
   neighboring router utilizing normal operation of the adjacency state
   machine.  This will ensure that any old adjacency on the neighboring
   router will be reinitialized.

   Upon receipt of an IIH with the SA bit set, the behavior described in
   Section 3.2.2 is followed.

   Upon starting, a router starts timer T2 for each LSPDB.

   For each interface (and in the case of a LAN circuit, for each
   level), when an adjacency reaches the UP state, the starting router
   starts a timer T1 and transmits an IIH containing the restart TLV
   with the RR bit clear and SA bit set.  Upon expiry of the timer T1,
   it is restarted and the IIH is retransmitted with both RR and SA bits
   set (only the RR bit has changed state from earlier IIHs).

   Upon receipt of an IIH with the RR bit set (regardless of whether or
   not the SA bit is set), the behavior described in Section 3.2.1 is
   followed.

   When an IIH is received by the starting router and the IIH contains a
   Restart TLV with the RA bit set (and on LAN circuits with a Restart
   Neighbor System ID that matches that of the local system), the
   receipt of the acknowledgement over that interface is noted.

   On a Point-to-Point link, receipt of an IIH not containing the
   Restart TLV is also treated as an acknowledgement, since it indicates
   that the neighbor is not restart capable.  Since the neighbor will
   have reinitialized the adjacency, this guarantees that SRMflags have
   been set on its database, thus ensuring eventual LSPDB
   synchronization.  However, since no CSNP is guaranteed to be received
   over this interface, the timer T1 is canceled immediately without
   waiting for a complete set of CSNPs.  Synchronization may therefore
   be deemed complete even though there are some LSPs that are held
   (only) by this neighbor (see Section 3.4).

   In the case of a LAN interface, receipt of an IIH not containing the
   Restart TLV is unremarkable since synchronization can still occur so
   long as at least one of the non-restarting neighboring routers on the
   LAN supports restart.  Therefore, T1 continues to run in this case.
   If none of the neighbors on the LAN are restart capable, T1 will
   eventually expire after the locally defined number of retries.  The
   usual operation of the update process will ensure that
   synchronization is eventually achieved.

   When BOTH a complete set of CSNPs (for each active level, in the case
   of a Point-to-Point circuit) and an acknowledgement have been
   received over the interface, the timer T1 is canceled.  Subsequent
   IIHs sent by the starting router have the RR and RA bits clear and
   the SA bit set in the Restart TLV.

   Timer T1 is canceled after some predetermined number of expirations
   (which MAY be 1).

   When the T2 timer(s) are canceled or expire, transmission of "normal
   IIHs" will begin.

3.3.3.  Multiple Levels



   A router that is operating as both a Level 1 and a Level 2 router on
   a particular interface MUST perform the above operations for each
   level.

   On a LAN interface, it MUST send and receive both Level 1 and Level 2
   IIHs and perform the CSNP synchronizations independently for each
   level.

   On a Point-to-Point interface, only a single IIH (indicating support
   for both levels) is required, but it MUST perform the CSNP
   synchronizations independently for each level.

3.4.  Database Synchronization



   When a router is started or restarted, it can expect to receive a
   complete set of CSNPs over each interface.  The arrival of the
   CSNP(s) is now guaranteed, since an IIH with the RR bit set will be
   retransmitted until the CSNP(s) are correctly received.

   The CSNPs describe the set of LSPs that are currently held by each
   neighbor.  Synchronization will be complete when all these LSPs have
   been received.

   When (re)starting, a router starts an instance of timer T2 for each
   LSPDB, as described in Section 3.3.1 or Section 3.3.2.  In addition
   to normal processing of the CSNPs, the set of LSPIDs contained in the
   first complete set of CSNPs received over each interface is recorded,
   together with their remaining lifetime.  In the case of a LAN
   interface, a complete set of CSNPs MUST consist of CSNPs received
   from neighbors that are not restarting.  If there are multiple
   interfaces on the (re)starting router, the recorded set of LSPIDs is
   the union of those received over each interface.  LSPs with a
   remaining lifetime of zero are NOT so recorded.

   As LSPs are received (by the normal operation of the update process)
   over any interface, the corresponding LSPID entry is removed (it is
   also removed if an LSP arrives before the CSNP containing the
   reference).  When an LSPID has been held in the list for its
   indicated remaining lifetime, it is removed from the list.  When the
   list of LSPIDs is empty and the timer T1 has been canceled for all
   the interfaces that have an adjacency at this level, the timer T2 is
   canceled.

   At this point, the local database is guaranteed to contain all the
   LSP(s) (either the same sequence number or a more recent sequence
   number) that were present in the neighbors' databases at the time of
   (re)starting.  LSPs that arrived in a neighbor's database after the
   time of (re)starting may or may not be present, but the normal
   operation of the update process will guarantee that they will
   eventually be received.  At this point, the local database is deemed
   to be "synchronized".

   Since LSPs mentioned in the CSNP(s) with a zero remaining lifetime
   are not recorded and those with a short remaining lifetime are
   deleted from the list when the lifetime expires, cancellation of the
   timer T2 will not be prevented by waiting for an LSP that will never
   arrive.

3.4.1.  LSP Generation and Flooding and SPF Computation



   The operation of a router starting, as opposed to restarting, is
   somewhat different.  These two cases are dealt with separately below.

3.4.1.1.  Restarting



   In order to avoid causing unnecessary routing churn in other routers,
   it is highly desirable that the router's own LSPs generated by the
   restarting system are the same as those previously present in the
   network (assuming no other changes have taken place).  It is
   important therefore not to regenerate and flood the LSPs until all
   the adjacencies have been reestablished and any information required
   for propagation into the local LSPs is fully available.  Ideally, the
   information is loaded into the LSPs in a deterministic way, such that
   the same information occurs in the same place in the same LSP (and
   hence the LSPs are identical to their previous versions).  If this
   can be achieved, the new versions may not even cause SPF to be run in
   other systems.  However, provided the same information is included in
   the set of LSPs (albeit in a different order, and possibly different
   LSPs), the result of running the SPF will be the same and will not
   cause churn to the forwarding tables.

   In the case of a restarting router, none of the router's own LSPs are
   transmitted, nor are the router's own forwarding tables updated while
   the timer T3 is running.

   Redistribution of inter-level information MUST be regenerated before
   this router's LSP is flooded to other nodes.  Therefore, the Level-n
   non-pseudonode LSP(s) MUST NOT be flooded until the other level's T2
   timer has expired and its SPF has been run.  This ensures that any
   inter-level information that is to be propagated can be included in
   the Level-n LSP(s).

   During this period, if one of the router's own (including
   pseudonodes) LSPs is received, which the local router does not
   currently have in its own database, it is NOT purged.  Under normal
   operation, such an LSP would be purged, since the LSP clearly should
   not be present in the global LSP database.  However, in the present
   circumstances, this would be highly undesirable, because it could
   cause premature removal of a router's own LSP -- and hence churn in
   remote routers.  Even if the local system has one or more of the
   router's own LSPs (which it has generated but not yet transmitted),
   it is still not valid to compare the received LSP against this set,
   since it may be that as a result of propagation between Level 1 and
   Level 2 (or vice versa), a further router's own LSP will need to be
   generated when the LSP databases have synchronized.

   During this period, a restarting router SHOULD send CSNPs as it
   normally would.  Information about the router's own LSPs MAY be
   included, but if it is included, it MUST be based on LSPs that have
   been received, not on versions that have been generated (but not yet
   transmitted).  This restriction is necessary to prevent premature
   removal of an LSP from the global LSP database.

   When the timer T2 expires or is canceled, indicating that
   synchronization for that level is complete, the SPF for that level is
   run in order to derive any information that is required to be
   propagated to another level, but the forwarding tables are not yet
   updated.

   Once the other level's SPF has run and any inter-level propagation
   has been resolved, the router's own LSPs can be generated and
   flooded.  Any own LSPs that were previously ignored, but that are not
   part of the current set of own LSPs (including pseudonodes), MUST
   then be purged.  Note that it is possible that a Designated Router
   change may have taken place and, consequently, the router SHOULD
   purge those pseudonode LSPs that it previously owned but that are now
   no longer part of its set of pseudonode LSPs.

   When all the T2 timers have expired or been canceled, the timer T3 is
   canceled, and the local forwarding tables are updated.

   If the timer T3 expires before all the T2 timers have expired or been
   canceled, this indicates that the synchronization process is taking
   longer than the minimum Holding Time of the neighbors.  The router's
   own LSP(s) for levels that have not yet completed their first SPF
   computation are then flooded with the overload bit set to indicate
   that the router's LSPDB is not yet synchronized (and therefore other
   routers MUST NOT compute routes through this router).  Normal
   operation of the update process resumes, and the local forwarding
   tables are updated.  In order to prevent the neighbor's adjacencies
   from expiring, IIHs with the normal interface value for the Holding
   Time are transmitted over all interfaces with neither RR nor RA set
   in the Restart TLV.  This will cause the neighbors to refresh their
   adjacencies.  The router's own LSP(s) will continue to have the
   overload bit set until timer T2 has expired or been canceled.

3.4.1.2.  Starting



   In the case of a starting router, as soon as each adjacency is
   established, and before any CSNP exchanges, the router's own zeroth
   LSP is transmitted with the overload bit set.  This prevents other
   routers from computing routes through the router until it has
   reliably acquired the complete set of LSPs.  The overload bit remains
   set in subsequent transmissions of the zeroth LSP (such as will occur
   if a previous copy of the router's own zeroth LSP is still present in
   the network) while any timer T2 is running.

   When all the T2 timers have been canceled, the router's own LSP(s)
   MAY be regenerated with the overload bit clear (assuming the router
   is not in fact overloaded, and there is no other reason, such as
   incomplete BGP convergence, to keep the overload bit set) and flooded
   as normal.

   Other LSPs owned by this router (including pseudonodes) are generated
   and flooded as normal, irrespective of the timer T2.  The SPF is also
   run as normal and the Routing Information Base (RIB) and Forwarding
   Information Base (FIB) updated as routes become available.

   To avoid the possible formation of temporary blackholes, the starting
   router sets the SA bit in the Restart TLV (as described in
   Section 3.3.2) in all IIHs that it sends.

   When all T2 timers have been canceled, the starting router MUST
   transmit IIHs with the SA bit clear.

4.  State Tables



   This section presents state tables that summarize the behaviors
   described in this document.  Other behaviors, in particular adjacency
   state transitions and LSP database update operations, are NOT
   included in the state tables except where this document modifies the
   behaviors described in [ISO10589] and [RFC5303].

   The states named in the columns of the tables below are a mixture of
   states that are specific to a single adjacency (ADJ suppressed, ADJ
   Seen RA, ADJ Seen CSNP) and states that are indicative of the state
   of the protocol instance (Running, Restarting, Starting, SPF Wait).

   Three state tables are presented from the point of view of a running
   router, a restarting router, and a starting router.

4.1.  Running Router



   +--------+-------------------------------------------+--------------+
   | Event  | Running                                   | ADJ          |
   |        |                                           | suppressed   |
   +========+===========================================+==============+
   | RX PR  | Set Planned Restart State                 |              |
   |        | Update Holding Time                       |              |
   |        | Send PA                                   |              |
   +--------+-------------------------------------------+--------------+
   | RX PR  | Clear Planned Restart State               |              |
   | clr    | Restore Holding Time to local value       |              |
   | and RR |                                           |              |
   | clr    |                                           |              |
   +--------+-------------------------------------------+--------------+
   | RX RR  | Maintain ADJ State                        |              |
   |        | Send RA                                   |              |
   |        | Set SRM, send CSNP (Note 1)               |              |
   |        | Update Holding Time,                      |              |
   |        | set Restart Mode (Note 2)                 |              |
   +--------+-------------------------------------------+--------------+
   | RX RR  | Clr Restart mode                          |              |
   | clr    |                                           |              |
   +--------+-------------------------------------------+--------------+
   | RX SA  | Suppress IS neighbor TLV in LSP(s)        |              |
   |        | Goto ADJ Suppressed                       |              |
   +--------+-------------------------------------------+--------------+
   | RX SA  |                                           | Unsuppress   |
   | clr    |                                           | IS neighbor  |
   |        |                                           | TLV in       |
   |        |                                           | LSP(s)       |
   |        |                                           | Goto Running |
   +--------+-------------------------------------------+--------------+

                          Table 1: Running Router

   Note 1:  CSNPs are sent by routers in accordance with item c in
            Section 3.2.1

   Note 2:  If Restart Mode clear

4.2.  Restarting Router



   +----------+-----------------+---------+---------+------------------+
   | Event    | Restarting      | ADJ     | ADJ     | SPF Wait         |
   |          |                 | Seen RA | Seen    |                  |
   |          |                 |         | CSNP    |                  |
   +==========+=================+=========+=========+==================+
   | Restart  | Send PR         |         |         |                  |
   | planned  |                 |         |         |                  |
   +----------+-----------------+---------+---------+------------------+
   | Planned  | Send PR clr     |         |         |                  |
   | restart  |                 |         |         |                  |
   | canceled |                 |         |         |                  |
   +----------+-----------------+---------+---------+------------------+
   | RX PA    | Proceed with    |         |         |                  |
   |          | planned restart |         |         |                  |
   +----------+-----------------+---------+---------+------------------+
   | Router   | Send IIH/RR     |         |         |                  |
   | restarts | ADJ Init        |         |         |                  |
   |          | Start T1, T2,   |         |         |                  |
   |          | T3              |         |         |                  |
   +----------+-----------------+---------+---------+------------------+
   | RX RR    | Send RA         |         |         |                  |
   +----------+-----------------+---------+---------+------------------+
   | RX RA    | Adjust T3       |         | Cancel  |                  |
   |          | Goto ADJ Seen   |         | T1      |                  |
   |          | RA              |         | Adjust  |                  |
   |          |                 |         | T3      |                  |
   +----------+-----------------+---------+---------+------------------+
   | RX CSNP  | Goto ADJ Seen   | Cancel  |         |                  |
   | set      | CSNP            | T1      |         |                  |
   +----------+-----------------+---------+---------+------------------+
   | RX IIH   | Cancel T1       |         |         |                  |
   | w/o      | (Point-to-point |         |         |                  |
   | Restart  | only)           |         |         |                  |
   | TLV      |                 |         |         |                  |
   +----------+-----------------+---------+---------+------------------+
   | T1       | Send IIH/RR     | Send    | Send    |                  |
   | expires  | Restart T1      | IIH/RR  | IIH/RR  |                  |
   |          |                 | Restart | Restart |                  |
   |          |                 | T1      | T1      |                  |
   +----------+-----------------+---------+---------+------------------+
   | T1       | Send IIH/normal | Send    | Send    |                  |
   | expires  |                 | IIH/    | IIH/    |                  |
   | nth time |                 | normal  | normal  |                  |
   +----------+-----------------+---------+---------+------------------+
   | T2       | Trigger SPF     |         |         |                  |
   | expires  | Goto SPF Wait   |         |         |                  |
   +----------+-----------------+---------+---------+------------------+
   | T3       | Set overload    |         |         |                  |
   | expires  | bit             |         |         |                  |
   |          | Flood local     |         |         |                  |
   |          | LSPs            |         |         |                  |
   |          | Update fwd      |         |         |                  |
   |          | plane           |         |         |                  |
   +----------+-----------------+---------+---------+------------------+
   | LSP DB   | Cancel T2 and   |         |         |                  |
   | Sync     | T3              |         |         |                  |
   |          | Trigger SPF     |         |         |                  |
   |          | Goto SPF wait   |         |         |                  |
   +----------+-----------------+---------+---------+------------------+
   | All SPF  |                 |         |         | Clear overload   |
   | done     |                 |         |         | bit              |
   |          |                 |         |         | Update fwd       |
   |          |                 |         |         | plane            |
   |          |                 |         |         | Flood local      |
   |          |                 |         |         | LSPs             |
   |          |                 |         |         | Goto Running     |
   +----------+-----------------+---------+---------+------------------+

                         Table 2: Restarting Router

4.3.  Starting Router



   +-------------+---------------------------+------------+------------+
   | Event       | Starting                  | ADJ Seen   | ADJ Seen   |
   |             |                           | RA         | CSNP       |
   +=============+===========================+============+============+
   | Router      | Send IIH/SA               |            |            |
   | starts      | Start T1 and T2           |            |            |
   +-------------+---------------------------+------------+------------+
   | RX RR       | Send RA                   |            |            |
   +-------------+---------------------------+------------+------------+
   | RX RA       | Goto ADJ Seen RA          |            | Cancel T1  |
   +-------------+---------------------------+------------+------------+
   | RX CSNP Set | Goto ADJ Seen CSNP        | Cancel T1  |            |
   +-------------+---------------------------+------------+------------+
   | RX IIH w no | Cancel T1 (Point-to-      |            |            |
   | Restart TLV | Point only)               |            |            |
   +-------------+---------------------------+------------+------------+
   | ADJ UP      | Start T1                  |            |            |
   |             | Send local LSPs with      |            |            |
   |             | overload bit set          |            |            |
   +-------------+---------------------------+------------+------------+
   | T1 expires  | Send IIH/RR and SA        | Send IIH/  | Send IIH/  |
   |             | Restart T1                | RR and SA  | RR and SA  |
   |             |                           | Restart T1 | Restart T1 |
   +-------------+---------------------------+------------+------------+
   | T1 expires  | Send IIH/SA               | Send IIH/  | Send IIH/  |
   | nth time    |                           | SA         | SA         |
   +-------------+---------------------------+------------+------------+
   | T2 expires  | Clear overload bit        |            |            |
   |             | Send IIH normal           |            |            |
   |             | Goto Running              |            |            |
   +-------------+---------------------------+------------+------------+
   | LSP DB Sync | Cancel T2                 |            |            |
   |             | Clear overload bit        |            |            |
   |             | Send IIH normal           |            |            |
   +-------------+---------------------------+------------+------------+

                          Table 3: Starting Router

5.  IANA Considerations



   This document defines the following IS-IS TLV that is listed in the
   "IS-IS TLV Codepoints" registry.

   +------+-------------+-----+-----+-----+-------+
   | Type | Description | IIH | LSP | SNP | Purge |
   +======+=============+=====+=====+=====+=======+
   | 211  | Restart TLV | y   | n   | n   | n     |
   +------+-------------+-----+-----+-----+-------+

                       Table 4

   IANA has updated the entry in registry to point to this document.

6.  Security Considerations



   Any new security issues raised by the procedures in this document
   depend upon the ability of an attacker to inject a false but
   apparently valid IIH, the ease/difficulty of which has not been
   altered.

   If the RR bit is set in a false IIH, neighbors who receive such an
   IIH will continue to maintain an existing adjacency in the UP state
   and may (re)send a complete set of CSNPs.  While the latter action is
   wasteful, neither action causes any disruption in correct protocol
   operation.

   If the RA bit is set in a false IIH, a (re)starting router that
   receives such an IIH may falsely believe that there is a neighbor on
   the corresponding interface that supports the procedures described in
   this document.  In the absence of receipt of a complete set of CSNPs
   on that interface, this could delay the completion of (re)start
   procedures by requiring the timer T1 to time out the locally defined
   maximum number of retries.  This behavior is the same as would occur
   on a LAN where none of the (re)starting router's neighbors support
   the procedures in this document and is covered in Sections 3.3.1 and
   3.3.2.

   If the SA bit is set in a false IIH, this could cause suppression of
   the advertisement of an IS neighbor, which could either continue for
   an indefinite period or occur intermittently with the result being a
   possible loss of reachability to some destinations in the network
   and/or increased frequency of LSP flooding and SPF calculation.

   If the PR bit is set in a false IIH, neighbors who receive such an
   IIH could modify the Holding Time of an existing adjacency
   inappropriately.  In the event of topology changes, the neighbor
   might also choose to not flood the topology updates and/or bring the
   adjacency down in the false belief that the forwarding plane of the
   router identified as the source of the false IIH is not currently
   processing announced topology changes.  This would result in
   unnecessary forwarding disruption.

   If the PA bit is set in a false IIH, a router that receives such an
   IIH may falsely believe that the neighbor on the corresponding
   interface supports the planned restart procedures defined in this
   document.  If such a router is planning to restart, it might then
   proceed to initiate a restart in the false expectation that the
   neighbor has updated its Holding Time as requested.  This may result
   in the neighbor bringing down the adjacency while the receiving
   router is restarting, causing unnecessary disruption to forwarding.

   The possibility of IS-IS PDU spoofing can be reduced by the use of
   authentication, as described in [RFC1195] and [ISO10589], and
   especially by the use of cryptographic authentication, as described
   in [RFC5304] and [RFC5310].

7.  Manageability Considerations



   These extensions that have been designed, developed, and deployed for
   many years do not have any new impact on management and operation of
   the IS-IS protocol via this standardization process.

8.  Normative References



   [ISO10589] International Organization for Standardization,
              "Information technology -- Telecommunications and
              information exchange between systems -- Intermediate
              System to Intermediate System intra-domain routeing
              information exchange protocol for use in conjunction with
              the protocol for providing the connectionless-mode network
              service (ISO 8473)", ISO/IEC 10589:2002, Second Edition,
              November 2002.

   [RFC1195]  Callon, R., "Use of OSI IS-IS for routing in TCP/IP and
              dual environments", RFC 1195, DOI 10.17487/RFC1195,
              December 1990, <https://www.rfc-editor.org/info/rfc1195>.

   [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>.

   [RFC5303]  Katz, D., Saluja, R., and D. Eastlake 3rd, "Three-Way
              Handshake for IS-IS Point-to-Point Adjacencies", RFC 5303,
              DOI 10.17487/RFC5303, October 2008,
              <https://www.rfc-editor.org/info/rfc5303>.

   [RFC5304]  Li, T. and R. Atkinson, "IS-IS Cryptographic
              Authentication", RFC 5304, DOI 10.17487/RFC5304, October
              2008, <https://www.rfc-editor.org/info/rfc5304>.

   [RFC5310]  Bhatia, M., Manral, V., Li, T., Atkinson, R., White, R.,
              and M. Fanto, "IS-IS Generic Cryptographic
              Authentication", RFC 5310, DOI 10.17487/RFC5310, February
              2009, <https://www.rfc-editor.org/info/rfc5310>.

   [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>.

   [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>.

Appendix A.  Summary of Changes from RFC 5306



   This document extends RFC 5306 by introducing support for signaling
   the neighbors of a restarting router that a planned restart is about
   to occur.  This allows the neighbors to be aware of the state of the
   restarting router so that appropriate action may be taken if other
   topology changes occur while the planned restart is in progress.
   Since the forwarding plane of the restarting router is maintained
   based upon the pre-restart state of the network, additional topology
   changes introduce the possibility that traffic may be lost if paths
   via the restarting router continue to be used while the restart is in
   progress.

   In support of this new functionality, two new flags have been
   introduced:

   PR -  Restart is planned

   PA -  Planned restart acknowledgement

   No changes to the post-restart exchange between the restarting router
   and its neighbors have been introduced.

Acknowledgements



   For RFC 5306, the authors acknowledged contributions made by Jeff
   Parker, Radia Perlman, Mark Schaefer, Naiming Shen, Nischal Sheth,
   Russ White, and Rena Yang.

   The authors of this updated document acknowledge the contribution of
   Mike Shand, coauthor of RFC 5306.

Authors' Addresses



   Les Ginsberg
   Cisco Systems, Inc.

   Email: ginsberg@cisco.com


   Paul Wells
   Cisco Systems, Inc.

   Email: pauwells@cisco.com