GENWiki

Premier IT Outsourcing and Support Services within the UK

User Tools

Site Tools


rfc:rfc8361

Internet Engineering Task Force (IETF) W. Hao Request for Comments: 8361 Y. Li Updates: 6325 Huawei Technologies Category: Standards Track M. Durrani ISSN: 2070-1721 Equinix

                                                              S. Gupta
                                                           IP Infusion
                                                                 A. Qu
                                                              MediaTec
                                                            April 2018
       Transparent Interconnection of Lots of Links (TRILL):
             Centralized Replication for Active-Active
      Broadcast, Unknown Unicast, and Multicast (BUM) Traffic

Abstract

 In Transparent Interconnection of Lots of Links (TRILL) active-active
 access, a Reverse Path Forwarding (RPF) check failure issue may occur
 when using the pseudo-nickname mechanism specified in RFC 7781.  This
 document describes a solution to resolve this RPF check failure issue
 through centralized replication.  All ingress Routing Bridges
 (RBridges) send Broadcast, Unknown Unicast, and Multicast (BUM)
 traffic to a centralized node with unicast TRILL encapsulation.  When
 the centralized node receives the BUM traffic, it decapsulates the
 packets and forwards them to their destination RBridges using a
 distribution tree established per the TRILL base protocol (RFC 6325).
 To avoid RPF check failure on an RBridge sitting between the ingress
 RBridge and the centralized replication node, some change in the RPF
 calculation algorithm is required.  RPF checks on each RBridge MUST
 be calculated as if the centralized node was the ingress RBridge,
 instead of being calculated using the actual ingress RBridge.  This
 document updates RFC 6325.

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/rfc8361.

Hao, et al. Standards Track [Page 1] RFC 8361 Centralized Replication for BUM Traffic April 2018

Copyright Notice

 Copyright (c) 2018 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. Introduction ....................................................2
 2. Conventions Used in This Document ...............................3
 3. Centralized Replication Solution Overview .......................4
 4. Frame Duplication from Remote RBridge ...........................6
 5. Local Forwarding Behavior on Ingress RBridge ....................6
 6. Loop Prevention among RBridges in an Edge Group .................8
 7. Centralized Replication Forwarding Process ......................9
 8. BUM Traffic Load-Balancing among Multiple Centralized Nodes ....10
 9. Coexisting with the CMT Solution (RFC 7783) ....................11
 10. Network Upgrade Analysis ......................................12
 11. TRILL Protocol Extensions .....................................13
    11.1. "R" and "C" Flag in the Nickname Flags APPsub-TLV ........13
 12. Security Considerations .......................................14
 13. IANA Considerations ...........................................14
 14. References ....................................................15
    14.1. Normative References .....................................15
    14.2. Informative References ...................................16
 Acknowledgments ...................................................17
 Authors' Addresses ................................................17

1. Introduction

 The IETF TRILL protocol [RFC6325] provides multipath data forwarding
 that is loop free and per-hop based with minimum configuration.
 TRILL uses IS-IS [RFC6165] [RFC7176] as its control plane routing
 protocol and defines a TRILL-specific header for user data.
 Customer Equipment (CE) devices can be multihomed to a set of edge
 RBridges forming an edge group where active-active service can be
 provided.  In that case, all of the uplinks from a CE are handled via
 a Local Active-Active Link Protocol (LAALP) [RFC7379] such as Multi-

Hao, et al. Standards Track [Page 2] RFC 8361 Centralized Replication for BUM Traffic April 2018

 Chassis Link Aggregation (MC-LAG) or Distributed Resilient Network
 Interconnect (DRNI) [IEEE802.1AX].  An active-active flow-based load-
 sharing mechanism can achieve better load-balancing and high
 reliability.  A CE device can be a Layer 3 (L3) end system by itself
 or a bridge switch through which L3 end systems access the TRILL
 campus.
 In active-active access, the pseudo-nickname solution in [RFC7781]
 can be used to avoid Media Access Control (MAC) flip-flop on remote
 RBridges.  The basic idea is to use a virtual RBridge (RBv) with a
 single pseudo-nickname to represent an edge group.  Any member
 RBridge of that edge group uses this pseudo-nickname rather than its
 own nickname as the ingress nickname when it injects TRILL data
 frames to the TRILL campus.  The use of the nickname solves the
 address flip-flop issue by setting the nickname learned by a remote
 RBridge to be the pseudo-nickname.  However, it introduces another
 issue of incorrect packet dropping as follows: When a pseudo-nickname
 is used by an edge RBridge as the ingress nickname to forward BUM
 traffic, any RBridges (RBn) sitting between the ingress RBridge and
 the distribution tree root will treat the traffic as if it were
 ingressed from the RBv.  If the same distribution tree is used by
 different edge RBridges of the same RBv, the traffic may arrive at
 some RBn from different ports.  Then, the Reverse Path Forwarding
 (RPF) check required by TRILL [RFC6325] fails, and the BUM traffic
 received on unexpected ports will be dropped by RBn.
 This document specifies a centralized replication solution for BUM
 traffic forwarding to resolve the issue of incorrect packet drop
 caused by the RPF check failure in the virtual RBridge case.  The
 basic idea is that all ingress RBridges send BUM traffic to a
 centralized node, which MUST be a distribution tree root, using
 unicast TRILL encapsulation.  When the centralized node receives the
 packets, it decapsulates and forwards them to their destination
 RBridges using a distribution tree established as per the TRILL base
 protocol.  This document updates [RFC6325]; per [RFC6325], multi-
 destination traffic is ingressed to a multi-destination TRILL data
 packet.  However, per this document, when using the centralized
 replication feature, multi-destination traffic is initially ingressed
 to a unicast TRILL data packet.

2. Conventions Used in This Document

 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.

Hao, et al. Standards Track [Page 3] RFC 8361 Centralized Replication for BUM Traffic April 2018

 The abbreviations and terminology in [RFC6325] are used herein with
 the following additions:
 BUM:           Broadcast, Unknown unicast, and Multicast
 CE:            Customer Equipment (as in [RFC7783]), as relates to a
                device (end station or bridge).  The device can be
                either physical or virtual equipment.
 Data Label:    VLAN or Fine-Grained Labeled (FGL) [RFC7172]
 DF:            Designated Forwarder [RFC7781]
 FGL:           Fine-Grained Label [RFC7172]
 LAALP:         Local Active-Active Link Protocol [RFC7379]
 MAC flip flop: A problem where the attachment point of a MAC address
                appears to a remote switch to keep changing.  See
                Section 3.3 of [RFC7379].
 MC-LAG:        Multi-Chassis Link Aggregation
 RPF:           Reverse Path Forwarding

3. Centralized Replication Solution Overview

 When an edge RBridge receives BUM traffic from a CE device, it uses
 unicast TRILL encapsulation instead of multicast encapsulation to
 send the packets to a centralized node.  The centralized node MUST be
 a distribution tree root.  Distribution tree roots are normally
 chosen to be high-capacity core RBridges with many high-bandwidth
 adjacencies.  This constraint makes it practical, as described below,
 to support centralized replication with only software changes to
 transit RBridges.
 The TRILL header of the unicast TRILL encapsulation contains an
 "ingress RBridge nickname" field and an "egress RBridge nickname"
 field [RFC6325].  If the ingress RBridge receives the BUM packet from
 a port that is in an active-active edge group using [RFC7781], it
 sets the ingress RBridge nickname to be the pseudo-nickname rather
 than its own nickname to avoid MAC flip-flop (see Section 3.3 of
 [RFC7379]) on remote RBridges.  The egress RBridge nickname is set to
 a special nickname of the centralized node that is used to
 differentiate the centralized replication purpose unicast TRILL
 encapsulation from a normal unicast TRILL encapsulation.  This
 special nickname is called an "R-nickname".

Hao, et al. Standards Track [Page 4] RFC 8361 Centralized Replication for BUM Traffic April 2018

 When the centralized RBridge receives a unicast TRILL-encapsulated
 packet with its R-nickname as the egress nickname, it decapsulates
 the packet.  Then, the centralized RBridge replicates and forwards
 the BUM packet to the packet's destination RBridges using one of the
 distribution trees established per the TRILL base protocol [RFC6325].
 It MUST use a distribution tree whose tree root is the centralized
 RBridge itself.  (An RBridge may be the root of more than one tree.)
 When the centralized RBridge forwards the BUM traffic, it simply
 sends it on the distribution tree as if it were a locally ingressed
 frame, except that the ingress nickname remains the same as that in
 the packet it received to ensure that the MAC address learning by all
 egress RBridges is bound to the pseudo-nickname.
 When the replicated packet is forwarded by each RBridge along the
 distribution tree starting from the centralized node, an RPF check is
 performed per [RFC6325].  For any RBridge sitting between the ingress
 RBridge and the centralized replication node, the incoming port of
 such a BUM packet should be the centralized-node-facing port, as the
 multicast traffic always comes from the centralized node in this
 solution.  However, the RPF port as the result of distribution tree
 calculation as specified in [RFC6325] will be the real ingress
 RBridge-facing port, as it uses the edge group's virtual RBridge as
 the ingress RBridge, so the RPF check will fail.
 To solve this problem, some change in the RPF test is required.  In
 this case, the RPF calculation on each RBridge should use the
 centralized node as the ingress RBridge for each tree for which that
 node is the root instead of the real ingress virtual RBridge to
 perform the calculation.  As a result, the RPF check will accept
 traffic on the centralized-node-facing port of the RBridge for multi-
 destination traffic.  This prevents incorrect frame drops by the RPF
 check.
 The change in the actual RPF check on a received multi-destination
 TRILL data packet is easy.  The RPF check from [RFC6325] is a check
 to see if a triple of {ingress nickname, tree, receiving RBridge
 port} is allowed.  (The tree is indicated by the nickname of its
 root, which is stored in the TRILL Header "egress nickname" field.)
 When determining the RPF check, if "ingress nickname" is using
 centralized replication (indicated by a C-nickname, see Section 9),
 then the check is based on distribution from the tree root.  If
 "ingress nickname" is not using centralized replication, then the
 check is based on distribution from the RBridge having the ingress
 nickname.
 To differentiate the centralized replication unicast TRILL
 encapsulation from normal unicast TRILL encapsulation, the R-nickname
 is introduced for centralized replication.  When the centralized node

Hao, et al. Standards Track [Page 5] RFC 8361 Centralized Replication for BUM Traffic April 2018

 receives unicast TRILL encapsulation traffic with the egress nickname
 R-nickname, it decapsulates the packet and then forwards the packet
 to the destination RBridges through a distribution tree for which it
 is the root by re-encapsulation as aforementioned.  In TRILL,
 RBridges can hold multiple nicknames, so the centralized RBridge
 simply obtains another nickname to use as the R-nickname.  The
 centralized RBridge or RBridges should announce their R-nickname to
 all TRILL campuses through the TRILL Link State PDU (LSP) extension
 specified in Section 11.

4. Frame Duplication from Remote RBridge

 Frame duplication may occur when a remote host sends a multi-
 destination frame to a local CE that has an active-active connection
 to the TRILL campus.  To avoid the local CE receiving multiple copies
 from a remote RBridge, the Designated Forwarder (DF) mechanism is
 supported for egress-direction multicast traffic.
 The DF election mechanism [RFC7781] allows only one port of one
 RBridge in an active-active group to forward multicast traffic from
 the TRILL campus to the local access side for each VLAN.  The basic
 idea of using DF is to elect one RBridge per VLAN from an edge group
 to be responsible for egressing the BUM traffic.  [RFC7781] describes
 the DF election mechanism among member RBridges involved in an edge
 group.
 If the DF election mechanism is used for frame-duplication
 prevention, access ports on an RBridge are categorized as one of
 three types: non-group, group DF port, and group non-DF port.  The
 last two types can be called group ports.  Each of the group ports is
 associated with a pseudo-nickname.  If consistent nickname allocation
 to edge group RBridges is used, it is possible that the same pseudo-
 nickname is associated with more than one port on a single RBridge.
 A typical scenario is that CE1 is connected to RB1 and RB2 by LAALP1,
 whereas CE2 is connected to RB1 and RB2 by LAALP2.  In order to
 conserve the number of pseudo-nicknames used, member ports for both
 LAALP1 and LAALP2 on RB1 and RB2 are all associated with the same
 pseudo-nickname.

5. Local Forwarding Behavior on Ingress RBridge

 When an ingress RBridge (RB1) receives BUM traffic from a local
 active-active connected CE (CE1) device, the traffic will be injected
 into the TRILL campus with TRILL encapsulation; it will be replicated
 and forwarded to all destination RBridges through central
 replication, including the ingress RBridge itself, along a TRILL

Hao, et al. Standards Track [Page 6] RFC 8361 Centralized Replication for BUM Traffic April 2018

 distribution tree.  To avoid the traffic looping back to the original
 sender CE, an ingress nickname of the CE group's pseudo-nickname is
 used for traffic filtering.
 However, if there are two CEs, say CE1 and CE2, connecting to the
 ingress RB1 and each associated with the same pseudo-nickname, RB1
 needs to locally replicate and forward to CE2, because another copy
 of the BUM traffic between CE1 and CE2 through the TRILL campus will
 be blocked by the traffic filtering.
 If CE1 and CE2 are not associated with the same pseudo-nickname, the
 copy of the BUM traffic between CE1 and CE2 through the TRILL campus
 won't be blocked by the traffic filtering.  To avoid duplicated
 traffic on receiver CE, there cannot be local replicated BUM traffic
 between these two CEs on ingress RB1.
 In summary, to ensure correct BUM traffic forwarding behavior for
 each CE, the local replication behavior on the ingress RBridge is as
 follows:
 1.  Replicate to the active-active group ports associated with the
     same pseudo-nickname as that associated with the incoming port.
 2.  Do not replicate to active-active group ports associated with
     other pseudo-nicknames.
 3.  Do not replicate to non-edge-group ports.
 The above local forwarding behavior on the ingress RBridge of RB1 can
 be called "centralized replication local forwarding behavior A".
 If ingress RBridge RB1 itself is the centralized replication node,
 BUM traffic injected by RB1 into the TRILL campus won't loop back to
 RB1.  In this case, the local forwarding behavior is called
 centralized replication local forwarding behavior B.  Behavior B on
 RB1 is as follows:
 1.  Local replication to the ports associated with the same pseudo-
     nickname as that associated with the incoming port.
 2.  Local replication to the group DF port associated with different
     pseudo-nicknames.  Do not replicate to group non-DF ports
     associated with different pseudo-nicknames.
 3.  Local replication to non-edge-group ports.

Hao, et al. Standards Track [Page 7] RFC 8361 Centralized Replication for BUM Traffic April 2018

6. Loop Prevention among RBridges in an Edge Group

 If a CE sends a BUM packet through a DF port to an ingress RBridge,
 that RBridge will forward that packet to all or a subset of the other
 RBridges that only have non-DF ports for that active-active group.
 Because BUM traffic forwarding to non-DF ports isn't allowed, in this
 case, the frame won't loop back to the CE.
 If a CE sends a BUM packet through a non-DF port to an ingress
 RBridge, say RB1, then RB1 will forward that packet to other RBridges
 that have a DF port for that active-active group.  In this case, the
 frame will loop back to the CE and the traffic split-horizon
 filtering mechanism is used to avoid looping back among RBridges in
 the edge group.
 This split-horizon mechanism relies on the ingress nickname field in
 the TRILL header to check if a packet's egress port belongs to the
 same active-active group as the packet's incoming port to the TRILL
 campus.
 When the ingress RBridge receives BUM traffic from an active-active
 connected CE device, the traffic will be sent through the TRILL
 campus with TRILL encapsulation to a centralized RBridge.  There it
 will be replicated and forwarded to its destination RBridges, which
 include the ingress RBridge itself, through a TRILL distribution
 tree.  If the same pseudo-nickname is used for two active-active
 access CEs as the ingress nickname, an egress RBridge can use that
 nickname to filter traffic forwarding to all local CEs.  In this
 case, the traffic between these two CEs goes through the local
 RBridge and another copy of the traffic from the TRILL campus is
 filtered.  If different ingress nicknames are used for two connecting
 CE devices, the access ports connecting to these two CEs should be
 isolated from each other.  The BUM traffic between these two CEs
 should go through the TRILL campus; otherwise, the destination CE
 connected to same RBridge with the sender CE will receive two copies
 of the traffic.

Hao, et al. Standards Track [Page 8] RFC 8361 Centralized Replication for BUM Traffic April 2018

7. Centralized Replication Forwarding Process

                           +-----------+
                           |   (RB5)   |
                           +-----------+
                                 |
                           +-----------+
                           |   (RB4)   |
                           +-----------+
                            |     |    |
                    --------      |     --------
                   |              |             |
                 +------+      +------+      +------+
                 |(RB1) |      |(RB2) |      | (RB3)|
                 +------+      +------+      +------+
                   *   |         *  |          * |  ^
                   *   |         *  |          * |   ^
                   *   ----------*-------------*--    ^
                   ***************************** |     ^
                   *                             |      ^
            LAALP1 *                      LAALP2 |       ^
               +------+                    +------+    +------+
               |  CE1 |                    | CE2  |    | CE3  |
               +------+                    +------+    +------+
                 Figure 1: TRILL Active-Active Access
 Note: The asterisk line, hyphen & vertical bar line, and circumflex
 line in this figure indicate the connection of the various CEs to the
 various RBs.
 Assuming the centralized replication solution is used in the example
 network of above Figure 1: RB5 is the distribution tree root and
 centralized replication node; CE1 and CE2 are active-active accessed
 to RB1, RB2, and RB3 through LAALP1 and LAALP2, respectively; and CE3
 is single-homed to RB3.  The RBridge's own nicknames of RB1 to RB5
 are nick1 to nick5, respectively.  RB1, RB2, and RB3 use the same
 pseudo-nickname for LAALP1 and LAALP2; that pseudo-nickname is
 P-nick.  The R-nickname on the centralized replication node of RB5 is
 S-nick.

Hao, et al. Standards Track [Page 9] RFC 8361 Centralized Replication for BUM Traffic April 2018

 The BUM traffic forwarding process from CE1 to CE2 and CE3 is as
 follows:
 1.  CE1 sends BUM traffic to RB3.
 2.  RB3 replicates and sends the BUM traffic to CE2 locally.  RB2
     also sends the traffic to RB5 using unicast TRILL encapsulation.
     In the TRILL Header, the ingress nickname is set as P-nick and
     the egress nickname is set as S-nick.
 3.  RB5 decapsulates the unicast TRILL data packet.  Then, it uses a
     distribution tree for which it is the root to forward the packet
     as a multi-destination TRILL data packet.  The egress nickname in
     the multi-destination TRILL Header is the nick5 and the ingress
     nickname is still P-nick.  If RB3 had sent the unicast to some
     nickname that was not an R-nickname, the packet would not be
     re-encapsulated.  If it is sent to an R-nickname that is not a
     tree root, it either will not be forwarded at all or, if it is
     re-encapsulated and forwarded, will be subject to incorrect
     pruning and will not be delivered to all of its intended
     recipients.
 4.  RB4 receives multicast TRILL traffic from RB5.  The incoming
     traffic port is the up port facing the distribution tree root.
     RB4's RPF check will be correct based on the changed RPF port
     calculation algorithm in this document.  After the RPF check is
     performed, it forwards the traffic to all other egress RBridges
     (RB1, RB2, and RB3).
 5.  RB3 receives multicast TRILL traffic from RB4.  It decapsulates
     the multi-destination TRILL data packet.  Because the ingress
     nickname of P-nick is equivalent to the nickname of local LAALPs
     connecting to CE1 and CE2, RB3 doesn't forward the traffic to CE1
     and CE2 to avoid a duplicated frame.  RB3 only forwards the
     packet to CE3.
 6.  RB1 and RB2 receive multicast TRILL traffic from RB4.  The
     forwarding process is similar to the process on RB3, i.e.,
     because the ingress nickname of P-nick is equivalent to the
     nickname of the local LAALPs connecting CE1 and CE2, they also
     don't forward the traffic to local CE1 and CE2.

8. BUM Traffic Load-Balancing among Multiple Centralized Nodes

 To support unicast TRILL encapsulation BUM traffic load-balancing,
 multiple centralized replication nodes can be deployed and the
 traffic can be spread over these nodes based on data label (VLAN or
 FGL).  Furthermore, if it was desirable for a centralized node to be

Hao, et al. Standards Track [Page 10] RFC 8361 Centralized Replication for BUM Traffic April 2018

 sent more of this BUM traffic, it could hold two or more R-nicknames.
 The share of BUM traffic it would receive would be proportional to
 the number of R-nicknames it held.
 Assuming there are k different R-nicknames held by centralized nodes
 in a TRILL campus, the VLAN-based (or FGL-based [RFC7172]) load-
 balancing algorithm used by an ingress active-active access RBridge
 is as follows:
 1.  All R-nicknames are ordered and numbered from 0 to k-1 in
     ascending order, treating the nicknames as unsigned 16-bit
     integers.
 2.  For data label ID m, choose the R-nickname whose index is given
     by (m mod k) as egress nickname for BUM traffic unicast TRILL
     encapsulation.
 For example, there are three R-Nicknames (RNs).  The RNs will be
 ordered RN0 to RN2.  Assuming there are five VLANs from VLAN ID1 to
 ID5 spreading among edge RBridges, the traffic in VLAN1 will go to
 RN1, VLAN2 will go to RN2, and so on.
 When an ingress RBridge participating in an active-active connection
 receives BUM traffic from a local CE, the RBridge decides which
 R-nickname to send the traffic to based on the VLAN-based load-
 spreading algorithm; thus, data-label-based load-balancing for the
 BUM traffic can be achieved using multiple centralized nodes/multiple
 R-nicknames.

9. Coexisting with the CMT Solution (RFC 7783)

                  +------+    +------+
                  |(RB6) |    |(RB7) |
                  +------+    +------+
    ------------------|-----------|----------------------
    |            |              |          |            |
 +------+    +------+       +------+    +------+     +------+
 |(RB1) |    |(RB2) |       |(RB3) |    |(RB4) |     |(RB5) |
 +------+    +------+       +------+    +------+     +------+
     |          |               |          |            |
     ------------               -------------------------
           |                               |
       +------+                         +------+
       |  CE1 |                         |  CE2 |
       +------+                         +------+
     Figure 2: CMT and Centralized Replication Coexisting Scenario

Hao, et al. Standards Track [Page 11] RFC 8361 Centralized Replication for BUM Traffic April 2018

 Both the centralized replication solution and the Coordinated
 Multicast Trees (CMT) solution from [RFC7783] rely on using pseudo-
 nicknames to avoid MAC flip-flop on remote RBridges.  These two
 solutions can coexist in a single TRILL campus.  Each solution can be
 selected by each active-active edge group of RBridges independently.
 As illustrated in Figure 2, RB1 and RB2 use CMT for CE1's active-
 active access; RB3, RB4, and RB5 use the centralized replication for
 CE2's active-active access.
 For the centralized replication solution, edge group RBridges MUST
 announce the local pseudo-nickname using the Nickname Flags APPsub-
 TLV with C flag set.  A nickname with the C flag set is called a
 "C-nickname".  A transit RBridge will perform the centralized
 replication-specific RPF check algorithm if it receives TRILL data
 packets with a C-nickname as the ingress nickname.
 An edge group using CMT [RFC7783] MUST NOT set the C flag on the
 pseudo-nickname it is using.  This is already mandatory behavior
 because any RBridge originating a Nickname Flags APPsub-TLV is
 required by [RFC7780] to set any flag bit it does not know about to
 zero.  If an edge RBridge using CMT [RFC7783] nevertheless set the
 C-bit for an edge group pseudo-nickname, it is very likely that BUM
 traffic encapsulated with that nickname as ingress would be
 incorrectly pruned early in its distribution and would, thus, reach
 few (possibly none) of its intended targets.  To avoid confusion, a
 pseudo-nickname MUST NOT be shared between a centralized replication
 edge group and a CMT-based edge group.

10. Network Upgrade Analysis

 Centralized nodes will typically need software and hardware upgrades
 to support centralized replication, which stitches together the TRILL
 unicast traffic decapsulation process and the process of normal TRILL
 multicast traffic forwarding along the distribution tree.
 Active-active connection edge RBridges will typically need software
 and hardware upgrades to support unicast TRILL encapsulation for BUM
 traffic; the process is similar to other head-end replication
 processes.
 Transit nodes typically need only a software upgrade to support the
 changed RPF port calculation algorithm.

Hao, et al. Standards Track [Page 12] RFC 8361 Centralized Replication for BUM Traffic April 2018

11. TRILL Protocol Extensions

 Two new flags, "R" and "C", are specified in the Nickname Flags
 APPsub-TLV [RFC7780].  A nickname with the R flag set is called an
 "R-nickname" and a nickname with the C flag set is called a
 "C-nickname".  The R-nickname is a specialized nickname attached to a
 centralized node to differentiate unicast TRILL-encapsulated BUM
 traffic from normal unicast TRILL traffic.  The C-nickname flag is
 set on the pseudo-nickname for each edge group that uses the
 centralized replication.  A C-nickname is a specialized pseudo-
 nickname for which transit RBridges perform a different RPF check
 algorithm for TRILL data packets with the C-nickname in the ingress
 nickname field.
 When active-active edge RBridges use centralized replication to
 forward BUM traffic, the R-nickname is used as the egress nickname
 and the C-nickname is used as ingress nickname in the TRILL header
 for the unicast TRILL encapsulation of BUM traffic.

11.1. "R" and "C" Flag in the Nickname Flags APPsub-TLV

 If this APPsub-TLV is being advertised by an RBridge that does not
 have the nickname appearing in the Nickname Flags APPsub-TLV, the R
 and C flag bits in the APPsub-TLV MUST be treated as if they were
 zero.  If an RBridge that is not a distribution tree root advertises
 an R-nickname, that nickname MUST NOT be treated as an R-nickname but
 rather as an ordinary nickname; that is, the R-nickname flag is
 ignored for all purposes if the nickname is held by an RBridge that
 is not a tree root.
            0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15
           +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+
           |   Nickname                                    |
           +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+
           |IN|SE|R | C|    RESV                           |
           +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+
                           NICKFLAG RECORD
 o  R = If the R flag is one, it indicates that the advertising TRILL
    switch holding Nickname is a centralized replication node, and
    Nickname is used as egress nickname for edge group RBridges to
    inject BUM traffic into the TRILL campus when the edge group
    RBridges use this centralized replication solution for active-
    active access.  If the R flag is zero, that nickname will not be
    used for that purpose.

Hao, et al. Standards Track [Page 13] RFC 8361 Centralized Replication for BUM Traffic April 2018

 o  C = If C flag is one, it indicates that the TRILL traffic with
    this nickname as an ingress nickname requires the special RPF
    check algorithm specified in Section 3.  If the C flag is zero,
    that nickname will not be used for that purpose.
 Due to errors or due to transient inconsistent LSPs when the link
 state database is converging after a configuration change or the
 like, it is possible for there to be inconsistent Nickname Flags
 APPsub-TLVs for the same nickname.  In this case, it is RECOMMENDED
 that the nickname be treated as if the R / C flag were set if any
 Nickname Flags APPsub-TLV for that nickname has the R / C flag set.

12. Security Considerations

 This document does not introduce any extra security risks.  A rogue
 RBridge that is a tree root can attract traffic by advertising an
 R-nickname.  However, this does not represent a substantial increase
 in risk as RBridges could cause problems in a number of other ways by
 advertising low-cost adjacencies or making themselves the highest
 priority tree root or the like.  In general, the protection against
 an untrusted device acting as an RBridge and wrecking havoc is to use
 IS-IS authentication [RFC5310] and configure and administer the TRILL
 campus so that only trusted RBridges have the authentication key.
 For general TRILL security considerations, see [RFC6325].  For
 security considerations related to pseudo-nickname active-active, see
 [RFC7781].

13. IANA Considerations

 IANA has assigned two bits in the Nickname Flags APPsubTLV flags for
 the R and C bits discussed in Section 11.1 and update the "NickFlags
 Bits" subregistry of the "Transparent Interconnection of Lots of
 Links (TRILL) Parameters" registry as follows:
            Bit  Mnemonic   Description          Reference
           ---  --------  --------------------  -----------
             2    R        Replication Nickname  [RFC8361]
             3    C        Special RPF Check     [RFC8361]

Hao, et al. Standards Track [Page 14] RFC 8361 Centralized Replication for BUM Traffic April 2018

14. References

14.1. Normative References

 [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>.
 [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>.
 [RFC6165]  Banerjee, A. and D. Ward, "Extensions to IS-IS for Layer-2
            Systems", RFC 6165, DOI 10.17487/RFC6165, April 2011,
            <https://www.rfc-editor.org/info/rfc6165>.
 [RFC6325]  Perlman, R., Eastlake 3rd, D., Dutt, D., Gai, S., and A.
            Ghanwani, "Routing Bridges (RBridges): Base Protocol
            Specification", RFC 6325, DOI 10.17487/RFC6325, July 2011,
            <https://www.rfc-editor.org/info/rfc6325>.
 [RFC7172]  Eastlake 3rd, D., Zhang, M., Agarwal, P., Perlman, R., and
            D. Dutt, "Transparent Interconnection of Lots of Links
            (TRILL): Fine-Grained Labeling", RFC 7172,
            DOI 10.17487/RFC7172, May 2014,
            <https://www.rfc-editor.org/info/rfc7172>.
 [RFC7176]  Eastlake 3rd, D., Senevirathne, T., Ghanwani, A., Dutt,
            D., and A. Banerjee, "Transparent Interconnection of Lots
            of Links (TRILL) Use of IS-IS", RFC 7176,
            DOI 10.17487/RFC7176, May 2014,
            <https://www.rfc-editor.org/info/rfc7176>.
 [RFC7780]  Eastlake 3rd, D., Zhang, M., Perlman, R., Banerjee, A.,
            Ghanwani, A., and S. Gupta, "Transparent Interconnection
            of Lots of Links (TRILL): Clarifications, Corrections, and
            Updates", RFC 7780, DOI 10.17487/RFC7780, February 2016,
            <https://www.rfc-editor.org/info/rfc7780>.
 [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>.

Hao, et al. Standards Track [Page 15] RFC 8361 Centralized Replication for BUM Traffic April 2018

14.2. Informative References

 [RFC7781]  Zhai, H., Senevirathne, T., Perlman, R., Zhang, M., and Y.
            Li, "Transparent Interconnection of Lots of Links (TRILL):
            Pseudo-Nickname for Active-Active Access", RFC 7781,
            DOI 10.17487/RFC7781, February 2016,
            <https://www.rfc-editor.org/info/rfc7781>.
 [RFC7379]  Li, Y., Hao, W., Perlman, R., Hudson, J., and H. Zhai,
            "Problem Statement and Goals for Active-Active Connection
            at the Transparent Interconnection of Lots of Links
            (TRILL) Edge", RFC 7379, DOI 10.17487/RFC7379, October
            2014, <https://www.rfc-editor.org/info/rfc7379>.
 [RFC7783]  Senevirathne, T., Pathangi, J., and J. Hudson,
            "Coordinated Multicast Trees (CMT) for Transparent
            Interconnection of Lots of Links (TRILL)", RFC 7783,
            DOI 10.17487/RFC7783, February 2016,
            <https://www.rfc-editor.org/info/rfc7783>.
 [IEEE802.1AX]
            IEEE, "IEEE Standard for Local and metropolitan area
            networks -- Link Aggregation", IEEE 802.1AX,
            DOI 10.1109/IEEESTD.2017.7888436, March 2017,
            <http://ieeexplore.ieee.org/document/7888436/>.

Hao, et al. Standards Track [Page 16] RFC 8361 Centralized Replication for BUM Traffic April 2018

Acknowledgments

 The authors wish to acknowledge the important contributions of Donald
 Eastlake, Hongjun Zhai, Xiaomin Wu, Liang Xia, and Francis Dupont.

Authors' Addresses

 Weiguo Hao
 Huawei Technologies
 101 Software Avenue,
 Nanjing 210012
 China
 Email: haoweiguo@huawei.com
 Yizhou Li
 Huawei Technologies
 101 Software Avenue,
 Nanjing 210012
 China
 Email: liyizhou@huawei.com
 Muhammad Durrani
 Equinix
 Email: mdurrani@equinix.com
 Sujay Gupta
 IP Infusion
 RMZ Centennial
 Mahadevapura Post
 Bangalore - 560048
 India
 Email: sujay.gupta@ipinfusion.com
 Andrew Qu
 MediaTec
 Email: laodulaodu@gmail.com

Hao, et al. Standards Track [Page 17]

/data/webs/external/dokuwiki/data/pages/rfc/rfc8361.txt · Last modified: 2018/04/13 03:59 by 127.0.0.1

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki