| < draft-yizhou-trill-tc-awareness-00.txt | draft-yizhou-trill-tc-awareness-01.txt > | |||
|---|---|---|---|---|
| TRILL Working Group Yizhou Li | TRILL Working Group Yizhou Li | |||
| Internet Draft Weiguo Hao | Internet Draft Weiguo Hao | |||
| Intended status: Standards Track Huawei Technologies | Intended status: Standards Track Huawei Technologies | |||
| Expires: November 2012 July 09, 2012 | Jon Hudson | |||
| Brocade | ||||
| Naveen Nimmu | ||||
| Broadcom | ||||
| Anoop Ghanwani | ||||
| DELL | ||||
| Expires: April 2013 October 21, 2012 | ||||
| Aware Spanning Tree Topology Change on RBridges | Aware Spanning Tree Topology Change on RBridges | |||
| draft-yizhou-trill-tc-awareness-00.txt | draft-yizhou-trill-tc-awareness-01.txt | |||
| Status of this Memo | Status of this Memo | |||
| This Internet-Draft is submitted in full conformance with the | This Internet-Draft is submitted in full conformance with the | |||
| provisions of BCP 78 and BCP 79. | provisions of BCP 78 and BCP 79. | |||
| Internet-Drafts are working documents of the Internet Engineering | Internet-Drafts are working documents of the Internet Engineering | |||
| Task Force (IETF), its areas, and its working groups. Note that | Task Force (IETF), its areas, and its working groups. Note that | |||
| other groups may also distribute working documents as Internet- | other groups may also distribute working documents as Internet- | |||
| Drafts. | Drafts. | |||
| skipping to change at page 1, line 32 | skipping to change at page 1, line 41 | |||
| and may be updated, replaced, or obsoleted by other documents at any | and may be updated, replaced, or obsoleted by other documents at any | |||
| time. It is inappropriate to use Internet-Drafts as reference | time. It is inappropriate to use Internet-Drafts as reference | |||
| material or to cite them other than as "work in progress." | material or to cite them other than as "work in progress." | |||
| The list of current Internet-Drafts can be accessed at | The list of current Internet-Drafts can be accessed at | |||
| http://www.ietf.org/ietf/1id-abstracts.txt | http://www.ietf.org/ietf/1id-abstracts.txt | |||
| The list of Internet-Draft Shadow Directories can be accessed at | The list of Internet-Draft Shadow Directories can be accessed at | |||
| http://www.ietf.org/shadow.html | http://www.ietf.org/shadow.html | |||
| This Internet-Draft will expire on January 9, 2009. | This Internet-Draft will expire on January 17, 2009. | |||
| Copyright Notice | Copyright Notice | |||
| Copyright (c) 2012 IETF Trust and the persons identified as the | Copyright (c) 2012 IETF Trust and the persons identified as the | |||
| document authors. All rights reserved. | document authors. All rights reserved. | |||
| This document is subject to BCP 78 and the IETF Trust's Legal | This document is subject to BCP 78 and the IETF Trust's Legal | |||
| Provisions Relating to IETF Documents | Provisions Relating to IETF Documents | |||
| (http://trustee.ietf.org/license-info) in effect on the date of | (http://trustee.ietf.org/license-info) in effect on the date of | |||
| publication of this document. Please review these documents | publication of this document. Please review these documents | |||
| carefully, as they describe your rights and restrictions with respect | carefully, as they describe your rights and restrictions with respect | |||
| to this document. Code Components extracted from this document must | to this document. Code Components extracted from this document must | |||
| include Simplified BSD License text as described in Section 4.e of | include Simplified BSD License text as described in Section 4.e of | |||
| the Trust Legal Provisions and are provided without warranty as | the Trust Legal Provisions and are provided without warranty as | |||
| described in the Simplified BSD License. | described in the Simplified BSD License. | |||
| Abstract | Abstract | |||
| When a local LAN running spanning tree protocol connecting to TRILL | ||||
| campus via more than one RBridge, there are several ways to perform | ||||
| loop avoidance. One of them illustrated by RFC6325 [RFC6325] A.3 was | ||||
| to make relevant ports on edge RBridges involving in spanning tree | ||||
| calculation. When edge RBridges are emulated as a single highest | ||||
| priority root, the local bridged LAN will be naturally partitioned | ||||
| after running spanning tree protocol. This approach achieves better | ||||
| link utilization and intra-VLAN load balancing in some scenarios. | ||||
| This document describes how the edge RBridges react to topology | ||||
| change occurring in bridged LAN in order to make the abovementioned | ||||
| spanning tree approach function correct. | ||||
| Table of Contents | Table of Contents | |||
| 1. Introduction ................................................ 2 | 1. Introduction ................................................ 3 | |||
| 1.1. Motivations ............................................ 4 | 1.1. Motivations ............................................ 5 | |||
| 2. Conventions used in this document............................ 5 | 2. Conventions used in this document ........................... 6 | |||
| 3. BPDU RBridge Channel......................................... 5 | 3. BPDU RBridge Channel......................................... 6 | |||
| 4. Operations .................................................. 6 | 4. Operations .................................................. 7 | |||
| 4.1. Sending BPDU using RBridge channel ..................... 7 | 4.1. Sending BPDU using RBridge channel ..................... 8 | |||
| 4.2. Receiving BPDU in RBridge channel ...................... 7 | 4.2. Receiving BPDU in RBridge channel ...................... 9 | |||
| 4.3. Informing the remote site............................... 8 | 4.3. Informing the remote site ............................. 10 | |||
| 5. Security Considerations...................................... 9 | 5. Security Considerations..................................... 11 | |||
| 6. IANA Considerations ......................................... 9 | 6. IANA Considerations ........................................ 12 | |||
| 7. References ................................................. 10 | 7. References ................................................. 12 | |||
| 7.1. Normative References................................... 10 | 7.1. Normative References................................... 12 | |||
| 7.2. Informative References................................. 10 | 7.2. Informative References................................. 13 | |||
| 8. Acknowledgments ............................................ 10 | 8. Acknowledgments ............................................ 13 | |||
| 1. Introduction | 1. Introduction | |||
| TRILL protocol [RFC6325] [RFC6439] described appointed forwarder | The TRILL protocol [RFC6325] provides the appointed forwarder | |||
| mechanism for loop avoidance in the scenario shown by Figure 1. Only | mechanism [RFC6439] for loop avoidance where, for part of the loop, | |||
| one of the RBridges is responsible for encapsulating/decapsulating a | the frame would be in TRILL encapsulated format, for example in the | |||
| given VLAN data frame on a link. Local bridged LAN runs normal | scenario shown by Figure 1. Only one of the RBridges is responsible | |||
| spanning tree protocol for loop avoidance. RBridge keeps track of the | for encapsulating/decapsulating a given VLAN's data frames on a link. | |||
| root bridge by listening to BPDUs received on the local port. This | Bridges in the local bridged LAN runs normal spanning tree protocol | |||
| information is reported per VLAN by the RBridge in its LSP and is | for local loop avoidance. RBridges keeps track of the root bridge by | |||
| used to detect the root change. Root change willtrigger the reset of | listening to BPDUs received on the local port. This information is | |||
| the inhibition timer of the appointed forwarder. When an RBridge | reported per VLAN by the RBridge in its LSP and is used to detect a | |||
| ceases to be appointed forwarder for a VLAN on a port, it sends | root bridge change. Root bridge changes trigger the reset of the | |||
| topology change BPDUs to purge the MAC table on local bridged LAN | inhibition timer of the appointed forwarder. When an RBridge ceases | |||
| switches. An RBridge never encapsulates or forwards any BPDU frame it | to be appointed forwarder for a VLAN on a port, it sends topology | |||
| receives [RFC6325]. | change BPDUs to purge the MAC table on local bridged LAN switches. An | |||
| RBridge conformable to [RFC6325] never encapsulates or forwards any | ||||
| [RFC6325] A.2 & A.3 presented the problems using the conventional | BPDU frame it receives. | |||
| approach shown in Figure 1. Native frames enter and leave a link via | ||||
| the link's appointed forwarder for the VLAN of the frame can cause | ||||
| congestion or suboptimal routing. Four methods was illustrated in | ||||
| [RFC6325] to solve the problem, | ||||
| 1. Use RBridge instead of conventional bridge | ||||
| 2. Re-arrange network topology | ||||
| 3. Carefully select the different appointed forwarders for VLANs if | ||||
| end stations on local bridged LAN can be separated into multiple | ||||
| VLANs | ||||
| 4. Configure the RBridges to be like one STP tree root in local | ||||
| bridged LAN. The RBridge ports that are connected to the bridged | ||||
| LAN send spanning tree configuration BPDUs. Then the bridged LAN | ||||
| is forced into partitions. Figure 2 shows its network topology. | ||||
| ------------------ | ------------------ | |||
| / \ | / \ | |||
| | Trill Network | | | Trill Network | | |||
| \ / | \ / | |||
| ------------------ | ------------------ | |||
| | | | | | | |||
| DRB| | | DRB| | | |||
| +------+ +------+ | +------+ +------+ | |||
| AF --->| RB1 | | RB2 | | AF --->| RB1 | | RB2 | | |||
| skipping to change at page 3, line 42 | skipping to change at page 4, line 5 | |||
| | | | | | | | | | | |||
| | | | | | | | | | | |||
| | | |<---blocked | | | | |<---blocked | | |||
| |Bridged | +----+ | | | |Bridged | +----+ | | | |||
| |LAN +-----| B3 |----+ | | |LAN +-----| B3 |----+ | | |||
| | +----+ | | | +----+ | | |||
| +---------------------------------------------+ | +---------------------------------------------+ | |||
| Figure 1 TRILL and bridged LAN topology | Figure 1 TRILL and bridged LAN topology | |||
| [RFC6325] A.2 & A.3 presented the problems using the conventional | ||||
| approach shown in Figure 1. Native frames enter and leave a link via | ||||
| the link's appointed forwarder for the VLAN of the frame can cause | ||||
| congestion or suboptimal routing. Four methods was illustrated in | ||||
| [RFC6325] to solve the problem, | ||||
| 1. Use RBridge instead of conventional bridge | ||||
| 2. Re-arrange network topology | ||||
| 3. Carefully select the different appointed forwarders for VLANs if | ||||
| end stations on local bridged LAN can be separated into multiple | ||||
| VLANs | ||||
| 4. Configure the RBridges to be like one STP tree root in local | ||||
| bridged LAN. The RBridge ports that are connected to the bridged | ||||
| LAN send spanning tree configuration BPDUs. Then the bridged LAN | ||||
| is forced into partitions. Figure 2 shows its network topology. | ||||
| Method 1 and 2 highly depends on the network topology and equipment | Method 1 and 2 highly depends on the network topology and equipment | |||
| types and therefore have very limited applicability. Method 3 and 4 | types and therefore have very limited applicability. Method 3 and 4 | |||
| have broader applicability. Method 4 is more applicable than method 3 | have broader applicability. Method 4 is more applicable than method 3 | |||
| if all end stations in bridged LAN are on the same VLAN or intra VLAN | if all end stations in bridged LAN are on the same VLAN or intra VLAN | |||
| load balancing is required to avoid per VLAN congestion and | load balancing is required to avoid per VLAN congestion and | |||
| suboptimal routing. The traffic discontinuity was caused by | suboptimal routing. The traffic discontinuity was caused by | |||
| inhibition timer setting in case of root change in method 3. Proper | inhibition timer setting in case of root change in method 3. Proper | |||
| timeout value has to be carefully chosen for tradeoff between | timeout value has to be carefully chosen for tradeoff between | |||
| unnecessary traffic continuity and potential loop. Method 4 | unnecessary traffic continuity and potential loop. Method 4 | |||
| eliminates the requirement of setting inhibition timer in case of | eliminates the requirement of setting inhibition timer in case of | |||
| root change. Therefore method 4 is considered as a very common | root change. Therefore method 4 is considered as a very common | |||
| practice in real deployment. | practice in real deployment. | |||
| 1.1. Motivations | ------------------ | |||
| / \ | ||||
| Bridged LAN may have topology change at any time. When RB1 & RB2 | | Trill Network | | |||
| serve as one single STP tree root, it is required that RB1 and RB2 | \ / | |||
| have to tunnel some BPDUs to help the bridged LAN convergence in | ------------------ | |||
| certain circumstances. Figure 2 is used to show such motivation in | ||||
| the given topology. | ||||
| ------------------ | ||||
| / \ | ||||
| | | | ||||
| | Trill Network | | ||||
| | | | ||||
| \ / | ||||
| ------------------ | ||||
| | | | | | | |||
| | | | | | | |||
| -----+-----------+---- | -----+-----------+---- | |||
| / +------+ +------+ \ <---highest pri | / +------+ +------+ \ <---emulated highest | |||
| | | RB1 | | RB2 | | root Bx | | | RB1 | | RB2 | | priority root Bx | |||
| ------------| +------+ +------+ /--------- | -------------| +------+ +------+ |--------- | |||
| | \-----+-----------+----- | | | \-----+-----------+-----/ | | |||
| | | | | | | | | | | |||
| | | | | | | | | | | |||
| | | | | | | | | | | |||
| | +----+ +----+ \|/ +----+ | | | +----+ +----+ \|/ +----+ | | |||
| | | B4 |-------| B1 |--- ---| B2 | | | | | B4 |-------| B1 |--- ---| B2 | | | |||
| | +----+ p1 +----+ /|\ +----+ | | | +----+ p1 +----+ /|\ +----+ | | |||
| | | | | | | | | | | | | |||
| | | blocked \|/ | | | | blocked \|/ | | |||
| | | - ----blocked | | | | - ----blocked | | |||
| |Bridged | /|\ | | |Bridged | /|\ | | |||
| |LAN | +----+ | | | |LAN | +----+ | | | |||
| | +-----| B3 |----+ | | | +-----| B3 |----+ | | |||
| | p1 +----+ p2 | | | p1 +----+ p2 | | |||
| ----------------------------------------------- | ----------------------------------------------- | |||
| Figure 2 RBs function as STP tree root topology | Figure 2 RBs function as STP tree root topology | |||
| 1.1. Motivations | ||||
| Bridged LANs may have topology changes at any time. When RB1 & RB2 | ||||
| serve as one single STP tree root as shown in Figure 2, it is | ||||
| required that RB1 and RB2 have to tunnel some BPDUs to help the | ||||
| bridged LAN convergence in certain circumstances. Figure 2 will be | ||||
| used to illustrate such motivation for rest of this subsection. | ||||
| RB1 & RB2 use the same bridge ID to emit spanning tree BPDUs as the | RB1 & RB2 use the same bridge ID to emit spanning tree BPDUs as the | |||
| highest priority root Bx. All bridges in LAN see RB1 and RB2 as a | highest priority root Bx. All bridges in LAN see RB1 and RB2 as a | |||
| single tree root. Therefore B1-B2 and B2-B3 links are blocked for | single tree root. Therefore B1-B2 and B2-B3 links are blocked for | |||
| loop avoidance after running spanning tree protocol. RB1 and RB2 will | loop avoidance by the spanning tree protocol. RB1 and RB2 will not | |||
| not receive TRILL-Hello from each other. Bridged LAN is logically | receive TRILL-Hello from each other. Bridged LAN is logically | |||
| partitioned into two parts. RB1 is DRB and AF for all VLANs in left | partitioned into two parts. RB1 is DRB and AF for all VLANs in left | |||
| partition and RB2 is DRB and AF in right partition. | partition and RB2 is DRB and AF in right partition. | |||
| If B1-B3 link fails for some reason, alternate port p2 on B3 will | If B1-B3 link fails for some reason, alternate port p2 on B3 will | |||
| send topology change (TC) BPDU to B2. B2-B3 link will start | send topology change (TC) BPDU to B2 as RSTP specifies [802.1D]. B2- | |||
| forwarding frames. TC BPDU is then sent from B2 to RB2. As RB2 never | B3 link will start forwarding frames. TC BPDU is then sent from B2 to | |||
| forwards BPDU frame to TRILL campus, left partition has no way to | RB2. As RB2 never forwards BPDU frame to TRILL campus, left partition | |||
| know the topology change. Therefore B4 will not able to correctly | has no way to know the topology change. Therefore B4 will not able to | |||
| purge the MACs learnt from port p1 for end stations connected to B3. | correctly purge the MACs learnt from port p1 for end stations | |||
| MAC table entry aging is the last resort in this case. In addition, a | connected to B3. MAC table entry aging is the last resort in this | |||
| remote end station may keep sending traffic to an end station | case. In addition, a remote end station may keep sending traffic to | |||
| connected to B3 via RB1-B1 which causes frame loss. Therefore some | an end station connected to B3 via RB1-B1 which causes frame loss. | |||
| mechanism must be used to purge the MACs learned both in the left | Therefore some mechanism must be used to purge the MACs learned both | |||
| partition of the bridged LAN and the remote Rbridges when topology | in the left partition of the bridged LAN and the remote Rbridges when | |||
| changes. This draft proposes to use RBridge channel [TRILLChannel] to | topology changes. This draft proposes to use RBridge channel | |||
| tunnel the TC BPDU to solve the issue. | [TRILLChannel] to tunnel the TC BPDU to solve the issue. | |||
| 2. Conventions used in this document | 2. Conventions used in this document | |||
| The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", | The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", | |||
| "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this | "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this | |||
| document are to be interpreted as described in RFC-2119 [RFC2119]. | document are to be interpreted as described in RFC-2119 [RFC2119]. | |||
| This document uses the terminologies defined in [RFC6325] along with | This document uses the terminologies defined in [RFC6325] along with | |||
| the following: | the following: | |||
| Root Bridge Group - A group of RBridges acting as a single tree root | Root Bridge Group - A group of RBridges acting as an emulated single | |||
| in a spanning tree instance in local bridged LAN | tree root in a spanning tree instance in local bridged LAN. The group | |||
| has at least two RBridges. | ||||
| 3. BPDU RBridge Channel | 3. BPDU RBridge Channel | |||
| A new channel protocol is defined to carry BPDU. | A new channel protocol is defined to carry BPDU. | |||
| Channel protocol code: TBD (BPDU) | Channel protocol code: TBD (BPDU) | |||
| | 0| 1| 2| 3| 4| 5| 6| 7| 8| 9|10|11|12|13|14|15| | | 0| 1| 2| 3| 4| 5| 6| 7| 8| 9|10|11|12|13|14|15| | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| | RBridge Channel | | | RBridge Channel | | |||
| | Header | | | Header | | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| | Reserved | | | Reserved | | |||
| | | | | | | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| . BPDU . | . BPDU . | |||
| . . | . . | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| Figure 3 RBridge Channel Format for BPDU | ||||
| 4. Operations | BPDU field is used to put the original BPDU frame. | |||
| Figure 3 shows TC BPDU tunneled from RB2 to RB1 using RBridge Channel. | ||||
| 4.1. Sending BPDU using RBridge channel | ||||
| In figure 3, when B1-B3 link fails, port p2 on B3 will start to send | The fields of TRILL header and inner Ethernet header SHOULD be set as | |||
| TC BPDU and go to forwarding state. RB2 receives TC BPDU from B2 | per [TRILLChannel] unless specified in this draft. | |||
| sequentially. RB2 encapsulates the TC BPDU in RBridge channel and | ||||
| sends it to RB1. | ||||
| Interested VLANs and Spanning Tree Roots Sub-TLV [RFC6326] carries | 4. Operations | |||
| spanning tree root bridge IDs seen for all ports for which the | ||||
| RBridge is the appointed forwarder for a VLAN. As RB1 and RB2 use the | ||||
| same bridge ID and that bridge ID is the spanning tree root, RB1 and | ||||
| RB2 are considered as in a root bridge group. | ||||
| When RBridge receives TC BPDU from an access port, it tunnels the | Figure 4 shows TC BPDU tunneled from RB2 to RB1 using RBridge Channel. | |||
| frame to all the other RBridges in the same root bridge group using | ||||
| RBridge channel protocol specified in section 3. Normally the number | ||||
| of RBridges in a root bridge group is limited, say 2 or 3; such | ||||
| tunneling is performed using TRILL unicast encapsulation. N members | ||||
| in a root bridge group results in N-1 unicast tunneled BPDU sent. In | ||||
| figure 3, RB2 knows RB1 is in the same root bridge group from LSP | ||||
| exchange; hence RB2 uses RB1's nickname as egress nickname and | ||||
| encapsulates the TC BPDU in RBridge channel. | ||||
| ------------------------ | ------------------------- | |||
| / \ | / \ | |||
| | | | / \ | |||
| | Trill Network | | | Trill Network | | |||
| | | | | | | |||
| | +---------------+ | | | +---------------+ | | |||
| | | 4.tunnel BPDU | | | | | 4.tunnel BPDU | | | |||
| | | in channel | | | | | in channel | | | |||
| \ | +-----------+ | / | \ | +-----------+ | / | |||
| \ | | | | / | \ | | | | / | |||
| \ ---|-|-----------|-|-- / | \ ---|-|-----------|-|-- / | |||
| / +-+ +--+ +--+ +-+ \ <---highest pri | / +-+ +--+ +--+ +-+ \ <--- emulated highest | |||
| | | RB1 | | RB2 | | root Bx | | | RB1 | | RB2 | | priority root Bx | |||
| ------------| +------+ +------+ /------------------ | ------------| +------+ +------+ |------------------ | |||
| | \-----+-----------+----- | | | \-----+-----------+-----/ | | |||
| | | | | | | | | | | |||
| | 5. TC BPDU | | | /|\ 3. TC BPDU | | | 5. TC BPDU | | | /|\ 3. TC BPDU | | |||
| | \|/ | | | | | | \|/ | | | | | |||
| | | | | | | | | | | |||
| | +----+ +----+ \|/ +----+ | | | +----+ +----+ \|/ +----+ | | |||
| | | B4 |-------| B1 |--- ---| B2 | | | | | B4 |-------| B1 |--- ---| B2 | | | |||
| | +----+ +----+ /|\ +----+ | | | +----+ +----+ /|\ +----+ | | |||
| | | | | | | | | | | | | |||
| | | blocked | | | | | blocked | | | |||
| | | |<---blocking to | | | | |<---blocking to | | |||
| | 1.link \|/ | forwarding | | | 1.link \|/ | forwarding | | |||
| | failure --> | | | | failure --> | | | |||
| | /|\ | | | | /|\ | | | |||
| | | | | | | | | | | |||
| | | +----+ p2 | /|\ | | | | +----+ p2 | /|\ | | |||
| | +--| B3 |-------+ | | | | +--| B3 |-------+ | | | |||
| |Bridged +----+ ---------+ | | |Bridged +----+ ---------+ | | |||
| |LAN 2. TC BPDU | | |LAN 2. TC BPDU | | |||
| | | | | | | |||
| -------------------------------------------------------| | -------------------------------------------------------| | |||
| Figure 3 Tunneled TC BPDU | Figure 4 Tunneled TC BPDU | |||
| 4.1. Sending BPDU using RBridge channel | ||||
| In figure 4, when B1-B3 link fails, alternate port p2 on B3 will | ||||
| start to send TC BPDU and go to forwarding state. RB2 receives TC | ||||
| BPDU from B2 sequentially. RB2 encapsulates the TC BPDU in RBridge | ||||
| channel and sends it to RB1. | ||||
| Interested VLANs and Spanning Tree Roots Sub-TLV [RFC6326] carries | ||||
| spanning tree root bridge IDs seen for all ports for which the | ||||
| RBridge is the appointed forwarder for a VLAN. As RB1 and RB2 use the | ||||
| same bridge ID and that bridge ID is the spanning tree root, RB1 and | ||||
| RB2 are considered as in a root bridge group. Static configuration of | ||||
| root bridge group is also allowed. | ||||
| When RBridge receives TC BPDU from an access port, it tunnels the | ||||
| frame to all the other RBridges in the same root bridge group using | ||||
| RBridge channel protocol specified in section 3. Normally the number | ||||
| of RBridges in a root bridge group is limited, say 2 or 3; such | ||||
| tunneling is performed using TRILL unicast encapsulation. N members | ||||
| in a root bridge group results in N-1 sequential unicast BPDU | ||||
| tunneled. In figure 4, RB2 knows RB1 is in the same root bridge group | ||||
| from LSP exchange; hence RB2 uses RB1's nickname as egress nickname | ||||
| and encapsulates the TC BPDU in RBridge channel. M bit in TRILL | ||||
| header SHOULD be 0. | ||||
| If TRILL Campus was partitioned temporarily in some unusual cases, | ||||
| RBridges in the same root bridge group may not reach each other. For | ||||
| instance, if RB2 was not able to reach RB1 through TRILL campus at | ||||
| some transition period due to network fault, RB1 would not receive | ||||
| the tunneled TC BPDU from RB2. Then the approach illustrated in this | ||||
| document will take effect again only after RB1 and RB2 connectivity | ||||
| via TRILL recovers from the network fault. | ||||
| It is possible to statically configure a root bridge group, especial | ||||
| when network is relatively small and stable. Therefore when an | ||||
| RBridge tunnels the TC BPDU to other members in the same root bridge | ||||
| group, it has to make sure the destination is reachable. | ||||
| If edges RBridges configured in the same root bridge group connect to | ||||
| separate TRILL campus intentionally, it is not recommended to use | ||||
| spanning tree partition mechanism and such root bridge group | ||||
| provisioning is normally considered as mis-configuration. | ||||
| 4.2. Receiving BPDU in RBridge channel | 4.2. Receiving BPDU in RBridge channel | |||
| When an RBridge receives a TC BPDU from RBridge channel, it | When an RBridge receives a TC BPDU from RBridge channel, it | |||
| determines the frame was sent from a RB in the same root bridge group. | determines the frame was sent from an RB in the same root bridge | |||
| Then RBridge decapsulates the frame and sends the original TC BPDU to | group. Then RBridge decapsulates the frame and sends the original TC | |||
| its local bridged LAN. TC BPDU will be flooded throughout in left | BPDU to its local bridged LAN. TC BPDU will be flooded throughout in | |||
| partition to clear MAC table in bridges. | the left partition to merge MAC table of bridges. | |||
| 4.3. Informing the remote site | 4.3. Informing the remote site | |||
| When local topology changes, the correspondence of end station and | When local topology changes, the correspondence of end station and | |||
| its attaching RBridge cached by remote RB may become invalid. The | its attaching RBridge cached by remote RB may become invalid. The | |||
| RBridges who is the appointed forwarder for the specified VLAN in | RBridges who is the appointed forwarder for the specified VLAN in | |||
| remote sites should be informed to clear the stale correspondence | remote sites should be informed to update the stale correspondence | |||
| table entry. | table entry. | |||
| When traffic is bi-directional, the remote RBridge will receive the | When traffic is bi-directional, the remote RBridge will receive the | |||
| data frames from the newly attached RBridge of the local end station. | data frames from the newly attached RBridge of the local end station. | |||
| The remote RBridge will update its MAC-Nickname correspondence table. | The remote RBridge will update its MAC-Nickname correspondence table | |||
| naturally though data frame learning. | ||||
| When traffic is uni-directional from the remote to local site or | When traffic is uni-directional from the remote to local site or | |||
| traffic from local to remote has to be triggered by traffic from | traffic from local to remote has to be triggered by traffic from | |||
| remote to local, remote RBridge will not receive the data frame from | remote to local, remote RBridge will not receive the data frame from | |||
| local RBridge to refresh its table. Then traffic discontinuity may | local RBridge to refresh its table. Then traffic discontinuity may | |||
| last for some time until the table entry aged out at remote RBridge. | last for some time until the table entry is aged out at the remote | |||
| RBridge. | ||||
| A lightweight method is to use RBridge channel to carry MAC purge | A lightweight method is to use RBridge channel to carry MAC purge | |||
| information. In Figure 3, When RB2 receives TC BPDU, it derives the | information. In Figure 4, When RB2 receives TC BPDU from B2, it | |||
| corresponding VLAN list. For example, if MSTP is used, RB2 will get | derives the corresponding VLAN list. For example, if MSTP is used, | |||
| the VLAN IDs in the same MSTP instance as TC BPDU. RB2 sends out MAC | RB2 will get the VLAN IDs in the same MSTP instance as TC BPDU. RB2 | |||
| purge information using RBridge channel with VLAN information and | sends out MAC purge information using RBridge channel with VLAN | |||
| RBidges nicknames in the same root bridge group. All remote RBridges | information and RBidges' nicknames in the same root bridge group. All | |||
| received MAC purge should clear its MAC-to-nickname correspondence | remote RBridges received MAC purge should clear its MAC-to-nickname | |||
| table for entries with the specified nicknames and VLAN IDs. If no | correspondence table for entries with the specified nicknames and | |||
| VLAN list is specified, the remote RBridges should clear the | VLAN IDs. If no VLAN list is specified, the remote RBridges should | |||
| correspondence in all VLANs relevant to the given nicknames. The MAC | clear the correspondence in all VLANs relevant to the given nicknames. | |||
| purge is recommended to send on the management VLAN in which all | The MAC purge is recommended to send on the management VLAN in which | |||
| RBridges joins. | all RBridges joins. | |||
| A new channel protocol code for MAC purge should be defined as | A new channel protocol code for MAC purge should be defined as | |||
| follows. | follows. | |||
| | 0| 1| 2| 3| 4| 5| 6| 7| 8| 9|10|11|12|13|14|15| | | 0| 1| 2| 3| 4| 5| 6| 7| 8| 9|10|11|12|13|14|15| | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| | RBridge Channel | | | RBridge Channel | | |||
| | Header | | | Header | | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| | Number of nicknames | nickname 1 | | | Number of nicknames | nickname 1 | | |||
| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | |||
| | nickname 1 | nickname 2 | | | nickname 1 | nickname 2 | | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| | nickname 2 | ... | | | nickname 2 | ... | | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| | ... | nickname n | | | ... | nickname n | | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| | nickname n | Num of VLAN blocks| | | nickname n | Num of VLAN blocks| | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| | Start.VLAN | | | | Start.VLAN | | | |||
| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | |||
| | End.VLAN | | | | End.VLAN | | | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| | Other Start/End VLAN list ... | | | Other Start/End VLAN list ... | | |||
| +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+ | |||
| Figure 5 RBridge Channel Format for Purge | ||||
| Number of nicknames: number of the following nicknames, which will be | ||||
| used by the receivers to purge their relevant MAC-to-nickname | ||||
| correspondence table entries. | ||||
| Num of VLAN blocks: number of the following VLAN block. A VLAN block | ||||
| is specified by a start and an end VLAN IDs. When start and end VLAN | ||||
| IDs are the same, it implies only one VLAN ID is in the block. When | ||||
| number of VLAN block is 0, it implies no VLAN ID is specified. | ||||
| For any nickname x specified and any VLAN y specified in this TLV, | ||||
| the receivers should purge MAC-to-nickname correspondence table | ||||
| entries with (any-MAC, VLAN-y, nickname-x). When number of VLAN block | ||||
| is 0, the receivers should purge entries with (any-MAC, any-VLAN, | ||||
| nickname-x). | ||||
| 5. Security Considerations | 5. Security Considerations | |||
| This document does not change the general RBridge security | This document does not change the general RBridge security | |||
| considerations of the TRILL base protocol and TRILL RBridge Channel. | considerations of the TRILL base protocol and TRILL RBridge Channel. | |||
| See Section 6 of [RFC6325] and section 7 of [TRILLChannel]. | See Section 6 of [RFC6325] and section 7 of [TRILLChannel]. | |||
| Forged TC BPDU may trigger RBridge continuously sending tunneled BPDU | Forged TC BPDU may trigger RBridges continuously sending tunneled | |||
| and MAC purge. It may cause denial-of-service in TRILL campus. | BPDU and MAC purges. It may cause denial-of-service in TRILL campus. | |||
| Similar as the traditional bridged LAN running spanning tree, it is | Similar as the traditional bridged LAN running spanning tree, it is | |||
| suggested to monitor the receiving rate of TC BPDU on bridged LAN | suggested to monitor the receiving rate of TC BPDU on bridged LAN | |||
| facing port of RBridges. If the receiving rate is beyond the | facing port of RBridges. If the receiving rate is beyond the | |||
| threshold, RBridge should only process and tunnel the TC BPDU in the | threshold, RBridge should only process and tunnel the TC BPDU in the | |||
| configured rate. | configured rate. | |||
| 6. IANA Considerations | 6. IANA Considerations | |||
| IANA is requested to allocate the new channel protocol codes as | IANA is requested to allocate the new channel protocol codes as | |||
| following. | following. | |||
| skipping to change at page 10, line 13 | skipping to change at page 12, line 30 | |||
| Channel protocol code X2: MAC purge | Channel protocol code X2: MAC purge | |||
| 7. References | 7. References | |||
| 7.1. Normative References | 7.1. Normative References | |||
| [RFC6325] Perlman, R., Eastlake 3rd, D., Dutt, D., Gai, S., and A. | [RFC6325] Perlman, R., Eastlake 3rd, D., Dutt, D., Gai, S., and A. | |||
| Ghanwani, "Routing Bridges (RBridges): Base Protocol | Ghanwani, "Routing Bridges (RBridges): Base Protocol | |||
| Specification", RFC 6325, July 2011. | Specification", RFC 6325, July 2011. | |||
| [RFC6326] Eastlake, D., Banerjee, A., Dutt, D., Perlman, R., and A. | ||||
| Ghanwani, "TRILL Use of IS-IS", RFC 6326, July 2011. | ||||
| [6326bis] Eastlake, D. et.al., ''Transparent Interconnection of Lots | [6326bis] Eastlake, D. et.al., ''Transparent Interconnection of Lots | |||
| of Links (TRILL) Use of IS-IS'', draft-eastlake-isis- | of Links (TRILL) Use of IS-IS'', draft-eastlake-isis- | |||
| rfc6326bis-07.txt, Work in Progress, December 2011. | rfc6326bis-07.txt, Work in Progress, December 2011. | |||
| [RFC6439] Eastlake, D. et.al., ''RBridge: Appointed Forwarder'', RFC | [RFC6439] Eastlake, D. et.al., ''RBridge: Appointed Forwarder'', RFC | |||
| 6439, November 2011. | 6439, November 2011. | |||
| [TRILLChannel] - Eastlake, D., V. Manral, Y. Li, S. Aldrin, D. Ward, | [TRILLChannel] - Eastlake, D., V. Manral, Y. Li, S. Aldrin, D. Ward, | |||
| "RBridges: RBridge Channel Support in TRILL", draft-ietf- | "RBridges: RBridge Channel Support in TRILL", draft-ietf- | |||
| trill-rbridge-channel, work in progress. | trill-rbridge-channel, work in progress. | |||
| skipping to change at page 11, line 23 | skipping to change at page 14, line 23 | |||
| Phone: +86-25-56625375 | Phone: +86-25-56625375 | |||
| Email: liyizhou@huawei.com | Email: liyizhou@huawei.com | |||
| Weiguo Hao | Weiguo Hao | |||
| Huawei Technologies | Huawei Technologies | |||
| 101 Software Avenue, | 101 Software Avenue, | |||
| Nanjing 210012 | Nanjing 210012 | |||
| China | China | |||
| Phone: +86-25-56623144 | Phone: +86-25-56623144 | |||
| Email: haoweiguo@huawei.com | Email: haoweiguo@huawei.com | |||
| John Hudson | ||||
| Brocade | ||||
| 120 Holger Way | ||||
| San Jose, CA 95134 | ||||
| USA. | ||||
| Email: jon.hudson@gmail.com | ||||
| Naveen Nimmu | ||||
| Broadcom | ||||
| 9th Floor, Building no 9, Raheja Mind space | ||||
| Hi-Tec City, Madhapur, | ||||
| Hyderabad - 500 081, INDIA | ||||
| Phone: +1-408-218-8893 | ||||
| Email: naveen@broadcom.com | ||||
| Anoop Ghanwani | ||||
| DELL | ||||
| 350 Holger Way | ||||
| San Jose, CA 95134 | ||||
| USA. | ||||
| Phone: +1-408-571-3500 | ||||
| Email: Anoop@duke.alumni.duke.edu | ||||
| End of changes. 33 change blocks. | ||||
| 142 lines changed or deleted | 206 lines changed or added | |||
This html diff was produced by rfcdiff 1.39p1. The latest version is available from http://tools.ietf.org/tools/rfcdiff/ | ||||