< 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/