CN103428088B - A kind of distribution of tree root, the method for Message processing and route-bridge - Google Patents

A kind of distribution of tree root, the method for Message processing and route-bridge Download PDF

Info

Publication number
CN103428088B
CN103428088B CN201210148395.0A CN201210148395A CN103428088B CN 103428088 B CN103428088 B CN 103428088B CN 201210148395 A CN201210148395 A CN 201210148395A CN 103428088 B CN103428088 B CN 103428088B
Authority
CN
China
Prior art keywords
group
gdrb
tree
routing bridge
message
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN201210148395.0A
Other languages
Chinese (zh)
Other versions
CN103428088A (en
Inventor
廖婷
翟洪军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZTE Corp
Original Assignee
ZTE Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ZTE Corp filed Critical ZTE Corp
Priority to CN201210148395.0A priority Critical patent/CN103428088B/en
Publication of CN103428088A publication Critical patent/CN103428088A/en
Application granted granted Critical
Publication of CN103428088B publication Critical patent/CN103428088B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)
  • Small-Scale Networks (AREA)

Abstract

The invention discloses a kind of distribution of tree root, the method for Message processing and route-bridges, apply in multi-link transparent interconnection (TRILL) network, the tree root distribution method includes:In route-bridge group, by one of route-bridge (RB) as the specified node (GDRB) of group, tree root is distributed for each RB in group according to choosing tree strategy, then notifies allocation result to each RB in group.Correspondingly, the route-bridge includes:Choosing tree unit, for when the RB is used as GDRB in route-bridge group, tree root to be distributed for each RB in group according to choosing tree strategy;Notification unit is notified for the allocation result that unit is each RB distribution tree roots to be set in the choosing to each RB in group.It can be abandoned after applying the present invention to avoid the mistake of message, and tree distribution is flexibly used, group member can specify node to carry out sufficient information sharing, flexible network configuration and the guarantee for providing network application by group.

Description

Method for tree root distribution and message processing and routing bridge
Technical Field
The present invention relates to the field of communications, and in particular, to a method for tree root distribution and packet processing and a routing bridge.
Background
TRILL (Transparent Interconnection over Lots of Links) is a standard for link layer (L2) networks recommended by the Internet Engineering Task Force (IETF). TRILL introduces an Intermediate System to Intermediate System (IS-IS) routing protocol to the L2 network, implemented as a control plane. In a TRILL network, a device running a TRILL protocol is called a routing Bridge (Router Bridge, abbreviated as RBridge or RB), where an RB connected to a terminal is called Edge RB (Edge RB, abbreviated as ERB), an Ingress RB is called Ingress RB, and an Egress RB is called Egress RB. At the entrance of the TRILL network, the original data frame of the End device (End Station) is encapsulated into a TRILL format (i.e. a TRILL header and an external frame header are added in front of the original data frame, the TRILL header mainly comprises nicknames and hop counts of the TRILL network entrance and exit routing bridges), so that the data frame forming the TRILL is transmitted in the TRILL network. The other routing bridges responsible for hop-by-hop transfer of TRILL data frames from Ingress RB to Egress RB are called transport routing bridges. The data message is decapsulated on the Egress RB, and is restored to an original message and transmitted to the terminal, and the Egress RB learns from which Ingress RB the original frame is imported and forms a corresponding MAC (Medium Access Control) information table: { D _ MAC, Ingress Nickname, … … }.
To avoid loops, at the boundary of the TRILL network, only one RBridge can provide services for an end system in any one VLAN (Virtual Local area network), where the RBridge is called a service provider of the end system, such as an VLAN-x designated Forwarder (AF) on a shared link. This provision, while effective in avoiding loops, also presents problems such as: after AF switching on the shared link, the change of Ingress Nickname in some MAC table entries on the remote Egress RB is brought; when an end system is connected to multiple rbridges through a point-to-point link Multi-homing (e.g., through a Multi-device link aggregation Group (MCLAG), more than two Ingress RB nodes send messages to the end system at the same source address, and for an RB where a destination address is located, messages sent from different Ingress RBs are received, a MAC information table containing the mapping relationship between the source MAC address and the node nickname is continuously refreshed, which causes a MAC address jump (flip-flop) problem. In order to avoid the problem of causing flip-flop of the MAC address on the remote RBridge, these links can only work in Active-Standby mode, which results in wasted bandwidth and difficulty in meeting the requirements of high throughput and high reliability of the high-performance data center.
For this purpose, the TRILL working Group proposes the concept of Routing Bridge Group (RBG) or Virtual routing bridge (RBv). The member nodes in a routing bridge group tend to have some of the same attributes, such as: the Local Area Network (LAN) and the Local Area Network (LAN) satisfy the MCLAG relationship, or the Area. Within an RBG, group members share a Nickname, referred to as the group Nickname, which is identified below by RBv. Each group member advertises its own group Nickname in the TRILL network to help other rbridges compute a path to RBv. When the members in the same group send messages, the ingress rb can use the Nickname of the group to encapsulate the messages. Since RBv is a logically divided group, not a specific real node, it can be considered as a node carried under the group members when a specific control plane is implemented. For the data plane, there may be a case where the RBv packet encapsulated from the node of the group member RB1 goes to the far end, and the packet replied to RBv from the far end is returned to the group member RB2, and for RB2, the reply may not be recognized, which may result in a case where the packet is discarded on RB2, and the RB1 cannot always receive the reply packet.
In particular, in the multicast case, the problem of tree selection and distribution of group members is also involved, specifically: in the TRILL network, multicast packets are forwarded along a tree, and a loop is avoided by using Reverse Path Forwarding checking (RPF) and other technologies. RPF means that for the routing bridge RB1, multicast packets from another routing bridge RB2 can only be received from one port in a given tree. If RB1 receives multicast packets from other ports outside this interface, which are sent from RB2 along this tree, then the RPF check is considered to fail, and the packets are discarded. By advertising which distribution trees to use by itself, a routing bridge may inform other bridges to calculate RPF information for it. When a plurality of trees are available in the routing bridge, the tree with the root closest to the routing bridge is selected by default when the message is forwarded, so that the forwarding efficiency is improved. Since the RPF is simple in that a leaf node is considered to be hung on only one parent node in a tree, if different group members are simultaneously encapsulated with RBv, RBv the leaf is considered to be hung behind the member nodes at the same time. If two group members simultaneously announce RBv on the same tree, it is equivalent to RBv that the two group members are simultaneously hung behind two father nodes and can not pass the RPF check. Therefore, Tree allocation is needed when RBv encapsulation is used for group members in the same group, and when a group member RB1 selects Tree1 for packet forwarding by RBv, other group member nodes cannot use RBv encapsulation on the Tree 1. The problem of RBv encapsulation versus tree selection is thus involved, i.e. group members are required to have to assign different trees when using RBv encapsulation.
Disclosure of Invention
The invention aims to provide a method for tree root distribution and message processing and a routing bridge, which are used for solving the problems of message error discarding and tree distribution in group members.
In order to solve the above problem, the present invention provides a method for tree root distribution, which is applied in a TRILL network, and comprises:
in the routing bridge group, one Routing Bridge (RB) is used as a group designation node (GDRB), a tree root is distributed to each RB in the group according to a tree selection strategy, and then the distribution result is notified to each RB in the group.
Further, the air conditioner is provided with a fan,
the GDRB in the routing bridge group is configured by the designation of an administrator, or each RB in the routing bridge group is elected from all RBs in the routing bridge group according to the same election rule.
Further, the air conditioner is provided with a fan,
under the condition that the GDRB is configured by the specification of an administrator, when the GDRB judges that the GDRB is configured as the GDRB, the GDRB sends GDRB identification messages for indicating that the GDRB is the GDRB to other RBs in the group;
and after receiving the GDRB identification message, the RB in the group is known to be selected as the GDRB.
Further, the air conditioner is provided with a fan,
the GDRB is elected from all RBs in the routing bridge group according to the same election rule by each RB in the routing bridge group, and specifically includes:
before the GDRB identification message is not received, selecting one RB from all RBs in the routing bridge group as the GDRB according to the same selection rule.
Further, the air conditioner is provided with a fan,
in the link aggregation group network, the election rule is that the RB with the highest priority of the root of the tree is preferentially elected in the group as the GDRB; or,
in the local area network, the election rule is that the RB with the highest priority of the interface in the group is preferentially elected as the GDRB; or,
in the region, the election rule is that one RB is selected as a GDRB according to the Nickname of each RB in the group.
Further, the air conditioner is provided with a fan,
the allocating tree roots for each RB in the group according to the tree selection strategy specifically includes:
the GDRB sorts all RBs in the group, and then tree roots are sequentially distributed to all the RBs in the group according to the sorting sequence; for each RB in the group, the GDRB selects a tree root with the shortest link cost or the largest link bandwidth from all currently unallocated tree roots to allocate to the RB.
Further, the method further comprises:
when a part of RB in the routing bridge group detects that distribution tree information to be used is configured on the equipment, the RB sends the distribution tree information to a GDRB in the routing bridge group;
the allocating tree roots for each RB in the group according to the tree selection strategy specifically includes:
after receiving the configured distribution tree information sent by other RBs in the group, the GDRB correspondingly preferentially allocates the tree to the RB; for each RB which is not configured with distribution tree information in the group, the GDRB selects a tree root with the shortest link cost or the largest link bandwidth with the RB from all currently unallocated tree roots to allocate to the RB.
Further, the air conditioner is provided with a fan,
after receiving the configured distribution tree information sent by other RBs in the group, the GDRB correspondingly preferentially allocates the tree to the RB, specifically including:
after receiving the configured distribution tree information sent by other RBs in the group, the GDRB preferentially allocates the preconfigured distribution tree to an RB with a high tree root priority for two or more RBs with the same preconfigured distribution tree information, and selects a tree root with the shortest link cost or the largest link bandwidth from all currently unallocated tree roots to allocate to the other RBs.
Further, the air conditioner is provided with a fan,
the notifying of the allocation result to each RB in the group specifically includes:
in a local area network, sending the distribution result to each RB in a group through a HELLO message in a TRILL network, or sending the distribution result to each RB in the group through a terminal address interaction protocol (ESADI) extension, or flooding to the TRILL network through a link state data packet (LSP), or sending the distribution result to each RB in the group through an intra-portal link among members of a multi-device link aggregation group (MCLAG);
and in the point-to-point network, the distribution result is sent to each RB in the group through ESADI extension, or is flooded to a TRILL network through a link state data packet (LSP), or is sent to each RB in the group on an intra-portal link among MCLAG members.
In addition, the invention also provides a message processing method, which is applied to the TRILL network and comprises the following steps:
in the routing bridge group, one Routing Bridge (RB) is used as a group designation node (GDRB);
after a member Routing Bridge (RB) receives a message sent from a remote end, if the message is judged to be a message which can not be processed by the member Routing Bridge (RB) and the destination address of the message is the message of the routing bridge group, the message is sent to a GDRB in the group;
after receiving the message, the GDRB forwards the message to other member RBs except the member RB from which the message is sent, if determining that the message is not a response message corresponding to a message sent by itself and encapsulated using the Nickname of the routing bridge group.
Further, the air conditioner is provided with a fan,
the GDRB in the routing bridge group is configured by the designation of an administrator, or each RB in the routing bridge group is elected from all RBs in the routing bridge group according to the same election rule.
Further, the air conditioner is provided with a fan,
under the condition that the GDRB is configured by the specification of an administrator, when the GDRB judges that the GDRB is configured as the GDRB, the GDRB sends GDRB identification messages for indicating that the GDRB is the GDRB to other RBs in the group;
and after receiving the GDRB identification message, the RB in the group is known to be selected as the GDRB.
Further, the air conditioner is provided with a fan,
the GDRB is elected from all RBs in the routing bridge group according to the same election rule by each RB in the routing bridge group, and specifically includes:
before the GDRB identification message is not received, selecting one RB from all RBs in the routing bridge group as the GDRB according to the same selection rule.
Further, the air conditioner is provided with a fan,
in the link aggregation group network, the election rule is that the RB with the highest priority of the root of the tree is preferentially elected in the group as the GDRB; or,
in the local area network, the election rule is that the RB with the highest priority of the interface in the group is preferentially elected as the GDRB; or,
in the region, the election rule is that one RB is selected as a GDRB according to the Nickname of each RB in the group.
Correspondingly, the invention provides a routing bridge, which is applied to a TRILL network and comprises the following components:
a tree selection unit, configured to allocate a tree root to each Routing Bridge (RB) in a group according to a tree selection policy when the RB serves as a group designation node (GDRB) in a routing bridge group;
and the notification unit is used for notifying the distribution result of the tree root distributed for each RB by the tree selection unit to each RB in the group.
Further, the air conditioner is provided with a fan,
and the notification unit is further configured to send a GDRB identity packet indicating that the routing bridge is a GDRB to other RBs in the group when determining that the routing bridge is configured as the GDRB.
Further, the routing bridge further comprises:
and the election unit is used for electing one RB from all RBs in the routing bridge group where the routing bridge is located as the GDRB according to an election rule before the GDRB identification message is not received.
Further, the air conditioner is provided with a fan,
the election unit elects the GDRB according to the following election rules:
in the link aggregation group network, the election rule is that the RB with the highest priority of the root of the tree is preferentially elected in the group as the GDRB; or,
in the local area network, the election rule is that the RB with the highest priority of the interface in the group is preferentially elected as the GDRB; or,
in the region, the election rule is that one RB is selected as a GDRB according to the Nickname of each RB in the group.
Further, the air conditioner is provided with a fan,
the tree selecting unit is configured to allocate tree roots for each RB in the group according to a tree selecting policy, and specifically includes:
the tree selecting unit is used for sequencing all RBs in the group and then sequentially distributing tree roots for the RBs in the group according to the sequencing sequence; and the tree selecting unit is used for selecting a tree root with the shortest link cost or the largest link bandwidth from all currently unallocated tree roots for each RB in the group to allocate to the RB.
Further, the air conditioner is provided with a fan,
the notification unit is further configured to send distribution tree information to be used to the GDRB in the routing bridge group where the routing bridge is located when detecting that the device has configured the distribution tree information to be used;
the tree selecting unit is configured to allocate tree roots for each RB in the group according to a tree selecting policy, and specifically includes:
the tree selecting unit is used for correspondingly and preferentially allocating the tree to the RB after receiving the configured distribution tree information sent by other RBs in the group; and for each RB which is not configured with the distribution tree information in the group, selecting the root with the shortest link cost or the largest link bandwidth from all the currently unallocated roots to allocate to the RB.
Further, the air conditioner is provided with a fan,
the tree selection unit is configured to, after receiving the configured distribution tree information sent by other RBs in the group, correspondingly preferentially allocate the tree to the RB, and specifically includes:
and the tree selecting unit is used for preferentially allocating the pre-configured distribution tree to the RB with the higher tree root priority for more than two RBs with the same pre-configured distribution tree information after receiving the configured distribution tree information sent by other RBs in the group, and selecting the tree root with the shortest link cost or the largest link bandwidth from all the currently unallocated tree roots to allocate to other RBs.
Further, the air conditioner is provided with a fan,
the notifying unit is configured to notify each RB in the group of the allocation result, and specifically includes:
the notification unit is configured to send, in the local area network, the allocation result to each RB in the group through a HELLO message in the TRILL network, or send the allocation result to each RB in the group through an end address interaction protocol (ESADI) extension, or flood the allocation result to the TRILL network through a Link State Packet (LSP), or send the allocation result to each RB in the group through an intra-portal link between members of a multi-device link aggregation group (MCLAG);
the notification unit is further configured to send, in the peer-to-peer network, the allocation result to each RB in the group through an ESADI extension, or flood to the TRILL network through a Link State Packet (LSP), or send the allocation result to each RB in the group on an intra-portal link between MCLAG members.
Correspondingly, the invention also provides a routing bridge, which is applied to the TRILL network and comprises the following components:
a receiving unit, configured to receive a message sent from a remote end or a message sent from another Routing Bridge (RB) in a routing bridge group where the remote end is located;
a judging unit, configured to judge whether the message is a message that cannot be processed by the receiving unit and has a destination address of the routing bridge group after the receiving unit receives the message sent by the remote end; the routing bridge is also used for judging whether a message from a far end, which is received by the receiving unit and sent by other RBs, is a response message which is sent by the receiving unit and corresponds to a message packaged by using the Nickname of the routing bridge group when the routing bridge is used as a group designated node (GDRB) in the routing bridge group where the routing bridge is located;
the sending unit is used for sending the message to the GDRB in the routing bridge group after the judging unit judges that the message sent from the remote end is the message which can not be processed by the sending unit and the destination address of the message is the routing bridge group; and when the judging unit judges that the message from the remote end sent by the other RB received by the receiving unit is not the response message corresponding to the message encapsulated by the Nickname of the routing bridge group sent by the judging unit when the routing bridge is used as the GDRB in the routing bridge group where the routing bridge is located, the judging unit is further configured to respectively forward the message to other member RBs except the member RB which sent the message.
Further, the air conditioner is provided with a fan,
and the sending unit is further configured to send a GDRB identity packet indicating that the routing bridge is a GDRB to other RBs in the group when determining that the routing bridge is configured as the GDRB.
Further, the air conditioner is provided with a fan,
and the election unit is used for electing one RB from all RBs in the routing bridge group where the routing bridge is located as the GDRB according to an election rule before the GDRB identification message is not received.
Further, the air conditioner is provided with a fan,
the election unit elects the GDRB according to the following election rules:
in the link aggregation group network, the election rule is that the RB with the highest priority of the root of the tree is preferentially elected in the group as the GDRB; or,
in the local area network, the election rule is that the RB with the highest priority of the interface in the group is preferentially elected as the GDRB; or,
in the region, the election rule is that one RB is selected as a GDRB according to the Nickname of each RB in the group.
The invention can avoid the error discarding of the message, flexibly apply the tree distribution, and the members in the group can fully share the information, flexibly configure the network and provide the guarantee of the network application through the designated nodes of the group.
Drawings
Fig. 1 is a topology diagram of a TRILL network according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for tree root distribution in a TRILL network according to an embodiment of the present invention;
fig. 3 is a diagram illustrating a structure of a TRILL network LAG group member in an exemplary application of the present invention;
FIG. 4 is a diagram illustrating a content of a notification message according to an exemplary embodiment of the present invention;
FIG. 5 is a diagram illustrating contents of another notification message in the first exemplary application of the present invention;
fig. 6 is a diagram of a TRILL network LAN group member structure in an application example two of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that the embodiments and features of the embodiments in the present application may be arbitrarily combined with each other without conflict.
In this embodiment, a method for performing packet processing in a TRILL network includes:
step 10: in the routing bridge group, one RB is used as a group designation node (GDRB);
after a certain member RB receives a message sent from a remote end, if it is determined that the message cannot be processed by itself and the destination address is a message of a routing bridge group where the message is located, the message is sent to a group designated node (distribution RB, hereinafter abbreviated as GDRB) in the group;
step 20: after receiving the message, the GDRB forwards the message to other member RBs except the member RB from which the message is sent, if determining that the message is not a response message corresponding to the message sent by itself and encapsulated by RBv.
As shown in fig. 1, RB1, RB2, and RB3 belong to the same routing bridge group, and after RBv1 encapsulated packets sent by RB1 reach remote RB5, RB5 replies response packets to RB 2. Since the RB2 cannot recognize the packet, such as an OAM (operations, administration, and Maintenance) packet, the RB2 preferentially sends the packet to the group designated node RB3 (here, it is assumed that RB3 is a group designated node in the group); after receiving the message sent by RB2 and addressed to RBv1, RB3 forwards the message to other member nodes in the group except for RB2, that is, to RB1 for processing if it is determined that the message does not need to be processed by itself; after receiving the response message, RB1 can correctly process the response message.
In addition, in this embodiment, a method for tree root distribution in a TRILL network, as shown in fig. 2, includes:
step 10: in the routing bridge group, one RB is used as a GDRB, and a tree root is distributed to each RB in the group according to a tree selection strategy. The group members are sorted according to tree root priority or other principles, and then tree selection calculation is sequentially carried out on all RBs in the group according to the tree selection strategy and the sorting sequence. Wherein: the tree selection strategy can be as follows: cost mini-principle or other strategies (such as bandwidth) and the like.
Assuming that the cost minimum optimization principle is used as a tree selection strategy, when a GDRB selects a tree for a group member, the GDRB selects a tree root according to a link cost between the group member and each tree root as a calculation basis, that is, when a tree is allocated to a certain group member, the tree with the shortest link cost (that is, the link cost of the tree root from the node of the group member is the smallest) is selected from all currently unallocated tree roots and allocated to the group member.
When a certain group member has manually configured a specific tree for use RBv in advance, the group member needs to send the pre-configured tree information to the GDRB in the group; after receiving the configuration information, the GDRB performs tree selection calculation again, and preferentially allocates the preconfigured tree for the group member; for the group members with conflict pre-configuration (for example, the same tree is pre-configured on more than two group members for packaging RBv), the tree is allocated to the group member with high tree root priority, the node with low tree root priority is additionally calculated and allocated, and the group members without pre-configured tree are calculated and allocated in the residual tree root according to the tree selection strategy.
Step 20: the GDRB informs the members in the group of the allocation result.
The distribution result can be carried in different messages according to different scenes. If the network is a LAN network, the network can be preferentially carried in HELLO messages of TRILL, or the TRILL network can be flooded by LSP (Link State Packet), or an extended ESADI (End State address distribution Information) can be used to carry notification messages, or the notification is carried on intra-port links (links between MCLAG devices) between MCLAG members; if the group is a point-to-point network group, such as an MCLAG mode group, the group can flood the TRILL network through the LSP, or carry an advertisement message by using extended ESADI, or advertise on an intra-portal link among MCLAG members. If the distribution result is announced by adopting an LSP (label switching path) mode, under the condition that the external node of the group can not identify the message, the RB (group member) sends out a corresponding Affinity TLV (Affinity Tag Length Value) according to the result after receiving the distribution result, wherein the Affinity TLV carries the distribution result when being locally used for RBv packaging; if it is known that other nodes outside the group can recognize the packet, the nodes outside the group can perform RPF calculation according to the received allocation result, and the group members only need to perform RBv encapsulation of the forwarding plane according to the tree allocated to the group members, and do not need to perform Affinity TLV announcement any more.
Thereafter, the group members perform RBv encapsulation on the data message according to the received tree allocation result, and optionally implement whether the control plane Affinity TLV advertisement needs to be performed. And other nodes outside the group carry out RBv RPF check according to the received distribution result of each member in the group.
It should be noted that, in the routing bridge group, the GDRB for performing tree distribution and forwarding the message may be the same RB or different RBs.
In specific implementation, the GDRB may be specified by an administrator or may be generated by a certain rule election, such as: in a LAG (Link Aggregation Group) network, an RB with the highest tree root priority may be preferentially elected within a Group as a GDRB; when more than two RBs with the highest tree root priority exist in the group, selecting the RB with the highest system id or the RB with the lowest system id as the GDRB according to the system id (for example, selecting the RB with the highest system id or the RB with the lowest system id from the RBs with the highest tree root priority, and of course, selecting the RB according to the system id by adopting other rules); when the system id of the RB selected according to a certain rule is the same, the selection can be further performed according to the Nickname; the RB with the highest interface priority can be preferentially elected within the group as a GDRB in the LAN network; when more than two RBs with the highest tree root priority exist in the group, selecting the RB with the highest system id or the RB with the lowest system id as the GDRB according to the system id (for example, selecting the RB with the highest system id or the RB with the lowest system id from the RBs with the highest tree root priority, and of course, selecting the RB according to the system id by adopting other rules); when the system id of the RB selected according to a certain rule is the same, the selection can be further performed according to the Nickname; within area, GDRB may be preferred according to Nickname.
Several examples of the invention are further described below.
Application example one: point-to-point LAG application scenarios
Since in the personal draft-tissa-TRILL-cmt-00 of the TRILL working group an Affinity TLV is proposed to announce out which trees the group members use when encapsulated using RBv. However, the scheme adopted by the document to tree distribution method is as follows: and performing modulo operation on the tree root sorting number and the sorting numbers of the group members, namely if n trees exist, sorting the n trees into 1-n, wherein j group members exist in the group, and the tree number distributed by the ith group member is i, i + j, i +2j.. i + xj in the n trees, wherein i + xj is less than n and less than i + (x +1) j. The allocation mode has no relation with the specific node and the physical position of the tree root node, and if the tree root allocated to the member RB1 is RBi and the tree root allocated to the member RB2 is RBj, when RB1 is close to RBj and RB2 is close to RBi in the physical position, the intermediate link bandwidth is wasted. And this approach does not meet this requirement if there are group members that want to configure a particular tree otherwise.
In view of this, in order to reasonably allocate the needed trees to the group members, it is necessary to perform computation negotiation among the group members according to the cost low-priority principle or other policies (such as bandwidth). Since congestion may exist on the link with the lowest cost, other policies such as bandwidth considerations may be selected for consideration. The group members perform the sorting calculation according to the principles of priority and the like, when there are many group members and there are many tree roots, each member node needs to calculate its own tree one by one, which means that all the group members calculate the trees that all the nodes will use, and the realization is complicated.
As shown in fig. 3, in the point-to-point network, RB 1-RBk are in a routing bridge group, the nickname of the group is RBv, if the tree root priorities of RB 1-RBk are sorted from high to low (or sorted according to other principles, for example, when the tree root priorities are consistent, sorted according to system id high and low, and sorted according to nickname size if system id is consistent) are RB 1-RBk, and it is assumed that RB1 is configured or selected as a GDRB from RB 1-RBk.
Suppose there are 3 trees in the current TRILL network, and the Tree roots of the 3 trees are RBm, RBj and RBi in order inside Tree id. If there is no configured designated tree on all group members, RB1 is the highest priority according to the COST minimum priority rule, then RB1 will preferentially select a tree for itself first. Calculating a first tree root with RBi of RB1 according to a tree selection strategy for selecting the tree root with the shortest distance from RB 1; then RB1 distributes tree roots for RB2, RBj is calculated from the rest of tree roots except RBi to be the tree root closest to RB2, and then RBj is selected to be distributed to RB 2; only the remaining last tree root RBm can be selected when selecting the calculation for RB 3.
If there are k +1 trees in the current TRILL network, the k +1 tree roots are RB1+ m, RB2+ m, RB3+ m, and RBk +1+ m in sequence (where m is a positive integer greater than k). When the tree is selected and allocated, the tree roots distributed to the RB1 to the RBk-2 by the RB1 are RB1+ m to RBk + m-2 in sequence, and when the tree is selected for the RBk-1, the tree roots which are not selected are RBk + m-1, RB k + m and RBk + m + 1. If the nearest to RBk-1 is RB k + m +1, then the tree root allocated to RBk-1 is RBk + m + 1; when a tree is selected for RBk, the nearest to the RBk is RB k + m-1, the root of the tree allocated to the RBk is RB k + m-1, the extra tree is circulated to RB1 for reselection, and calculation allocation is performed if the tree is calculated, namely two trees are allocated for RB 1.
After the allocation is completed, RB1 sends out the calculation result in a notification message, as shown in fig. 4, where the notification message carries the correspondence between the group members and the allocated tree numbers. Because this part of information is not complete and has no use for other intra-group nodes, it is only necessary to transmit the notification message in the group (for example, through intra-portal link transmission among group members or through adding an ESADI instance to all the intra-group members, and through ESADI, transmitting the corresponding information carrying the group members and the assigned tree number to the intra-group), after receiving the above assignment result, the group members send out the relevant AFF TLV to flood the TRILL network, thereby completing tree assignment and RPF check.
If the notification message carries RBv, the corresponding relationship among the group members and the tree numbers of the allocated trees, as shown in fig. 5, GDRB can select to flood the notification message into TRILL domain or group for transmission through intra-portal link, if the flooding can be performed by other nodes outside the extended information group, the group members do not need to send related AFF TLV, and the nodes outside the group can extract effective information for RPF check directly according to the notification message; if the nodes outside the group can not identify or only select to transmit through the intra-portal link in the group, the group members send out relevant AFF TLV flooding TRILL network after receiving the notification message, thereby completing tree distribution and RPF check.
Application example two: LAN application scenario
As shown in fig. 6, RB 1-RBk are in a routing bridge group, the nickname of the routing bridge group is RBv, if the tree root priorities of RB 1-RBk are ordered from high to low (or ordered according to other principles, for example, when the tree root priorities are consistent, they are ordered according to system id, if system id is consistent, they are ordered according to nickname of nickname), RB 1-RBk is defined, and if RB 1-RBk is listed as GDRB according to configuration or RB 2. According to the method of the application example, the RB2 performs tree selection calculation, and the tree distribution result is issued to other group members.
The notification information containing the assignment result can be selected to be carried in a HELLO packet on the LAN, and the packet structure is as shown in fig. 4, so that the rest nodes do not need to sense the existence of the packet, and the group member carries the assignment result in, for example, an Affinity TLV and sends out the assignment result after receiving the notification information. And the other external nodes carry out corresponding RPF check according to the message information. Of course, if the notification message shown in fig. 5 is sent, if it is desired that all the nodes outside the group can receive and identify the notification message, the LSP flooding may be selected, so that the group members may not send another message after receiving the notification message, and the other nodes outside the group can perform corresponding RPF check according to the received information. Of course, it is also allowable to use ESADI and intra-portal link for the transmission of the notification message in the LAN scenario.
Application example three: GDRB designation, election, and failure handling
The specification, election and failure processing of the GDRB in the application example one and the application example two may specifically be:
when the designated GDRB exists in the group, the GDRB can automatically set the election priority of the GDRB to be the highest according to an automatic election rule, and simultaneously sends out an identification message (which can be transmitted through an intra-portal link in the group) of the GDRB, other group members stop election after receiving the GDRB identification message sent by other members, and the GDRB is statically waited to release a distribution result. When other group members know that the GDRB is invalid (such as DBR is down), if other nodes in the group are configured with the GDRB, the node sets the election priority of the node to be the highest and sends GDRB identification messages, so that the nodes in the group know that the members take the GDRB role and the waiting distribution result. When the GDRB identification message is not received, the GDRB is not appointed in the group, and all members in the group can automatically elect according to a uniform election rule, namely: in the point-to-point network environment, election can be performed depending on election rules such as tree root priority level, system id size, nickname size, etc., in the LAN environment, GDRB election rules (i.e., election is performed according to interface priority, interface MAC, etc., and of course, the above-mentioned election rules in the point-to-point network can also be used) can be followed, and in short, each group member follows some unified rule. The selected GDRB sends out a GDRB identification message (which can be transmitted through an intra-portal link in the group) indicating that the GDRB is in the group, other group members do not select the GDRB after knowing that the members take the GDRB role, and the static waiting GDRB sends the distribution result to the local terminal. Therefore, once the GDRB failure is detected, a new GDRB is elected, the election of the new GDRB is carried out in a manner that the GDRB is elected according to configuration and the GDRB is elected according to election rules if the GDRB failure is not configured; and after the selected GDRB fails, continuously electing in other members in the group according to the principle, and when the members in the group receive a new GDRB identification message, indicating that the election is successful. When receiving multiple GDRB identification messages sent by different nodes at the same time, the multiple nodes sending the GDRB identification messages are elected according to a preset rule, a unique result is elected, the elected GDRB sends the GDRB identification messages, and after a distribution tree is distributed to members in a group, distribution results are sent periodically, so that updating and maintenance are carried out.
Application example four: tree assignment computation if there is a specified tree configuration on a group member
As shown in fig. 3, if the Used trees of RB1, RB2, and RBk are respectively configured as RBj, RBm, and RBi, and the notification of these Tree information is notified via Tree Used ID, the RPF check at this time is performed by nickname of each node. The tree assignment with RBv encapsulation calculated by the node is checked with RBv RPF, so there is no conflict. Whereas if the configuration is a tree-specific configuration exists when the tree is encapsulated with RBv, when a particular tree is manually configured for use RBv on a member node, the member node will send the configured distribution tree information to the GDRB. After the GDRB collects the configured distribution tree information, it recalculates, preferentially allocates the configured tree to the nodes configured with the distribution tree, and for the nodes with configuration conflict (for example, the same tree is configured on the two nodes for RBv encapsulation), the configured tree is occupied by the node with high tree root priority, and is allocated by calculation to the node with low tree root priority in the conflict. The member nodes needing calculation distribution can perform tree calculation distribution in the rest tree roots according to the calculation mode in the application example one and the tree selection strategy. If the subsequent nodes continue to sequentially select the tree root closest to the subsequent nodes from the unassigned tree roots, the only distribution of the tree on the group members can be realized under the condition of considering the distribution of the manually configured appointed tree as long as the trees distributed on the nodes of each member are not repeated.
Application example five: GDRB is responsible for OAM message transfer among group members
Assuming that RB1, RB2, and RB3 belong to the same routing bridge group, where RB3 is a GDRB, after a RBv encapsulated data packet sent by RB1 reaches remote RB5, RB5 finds that the packet encapsulation may have a problem, and needs to return an OAM error report to RBv. By querying the local table entry information, if it is found that the optimal route to RBv is the path between RB2, RB5 sends an OAM error report message back to RB2 through the path in the direction of RB 2. Because the RB2 does not send a message in the OAM error report description, which is sent by the RB1, when the RB2 receives the OAM error report and cannot identify it, it sends the OAM error report to the group designated node RB3 through an intra-portal link. After receiving the OAM error report whose destination address is RBv sent by RB2, RB3 first determines whether the packet needs to be processed, and if not, forwards the packet to other member nodes except the sender of the packet through intra-portal link, which is also RB1 in this example. After receiving the OAM error report, RB1 processes the report correctly.
Furthermore, in this embodiment, a routing bridge, applied in a TRILL network, includes:
a tree selection unit, configured to allocate a tree root to each Routing Bridge (RB) in a group according to a tree selection policy when the RB serves as a group designation node (GDRB) in a routing bridge group;
and the notification unit is used for notifying the distribution result of the tree root distributed for each RB by the tree selection unit to each RB in the group.
Preferably, the first and second liquid crystal films are made of a polymer,
and the notification unit is further configured to send a GDRB identity packet indicating that the routing bridge is a GDRB to other RBs in the group when determining that the routing bridge is configured as the GDRB.
Preferably, the routing bridge further comprises:
and the election unit is used for electing one RB from all RBs in the routing bridge group where the routing bridge is located as the GDRB according to an election rule before the GDRB identification message is not received.
Preferably, the first and second liquid crystal films are made of a polymer,
the election unit elects the GDRB according to the following election rules:
in the link aggregation group network, the election rule is that the RB with the highest priority of the root of the tree is preferentially elected in the group as the GDRB; or,
in the local area network, the election rule is that the RB with the highest priority of the interface in the group is preferentially elected as the GDRB; or,
in the region, the election rule is that one RB is selected as a GDRB according to the Nickname of each RB in the group.
Preferably, the first and second liquid crystal films are made of a polymer,
the tree selecting unit is configured to allocate tree roots for each RB in the group according to a tree selecting policy, and specifically includes:
the tree selecting unit is used for sequencing all RBs in the group and then sequentially distributing tree roots for the RBs in the group according to the sequencing sequence; and the tree selecting unit is used for selecting a tree root with the shortest link cost or the largest link bandwidth from all currently unallocated tree roots for each RB in the group to allocate to the RB.
Preferably, the first and second liquid crystal films are made of a polymer,
the notification unit is further configured to send distribution tree information to be used to the GDRB in the routing bridge group where the routing bridge is located when detecting that the device has configured the distribution tree information to be used;
the tree selecting unit is configured to allocate tree roots for each RB in the group according to a tree selecting policy, and specifically includes:
the tree selecting unit is used for correspondingly and preferentially allocating the tree to the RB after receiving the configured distribution tree information sent by other RBs in the group; and for each RB which is not configured with the distribution tree information in the group, selecting the root with the shortest link cost or the largest link bandwidth from all the currently unallocated roots to allocate to the RB.
Preferably, the first and second liquid crystal films are made of a polymer,
the tree selection unit is configured to, after receiving the configured distribution tree information sent by other RBs in the group, correspondingly preferentially allocate the tree to the RB, and specifically includes:
and the tree selecting unit is used for preferentially allocating the pre-configured distribution tree to the RB with the higher tree root priority for more than two RBs with the same pre-configured distribution tree information after receiving the configured distribution tree information sent by other RBs in the group, and selecting the tree root with the shortest link cost or the largest link bandwidth from all the currently unallocated tree roots to allocate to other RBs.
Preferably, the first and second liquid crystal films are made of a polymer,
the notifying unit is configured to notify each RB in the group of the allocation result, and specifically includes:
the notification unit is configured to send, in the local area network, the allocation result to each RB in the group through a HELLO message in the TRILL network, or send the allocation result to each RB in the group through an end address interaction protocol (ESADI) extension, or flood the allocation result to the TRILL network through a Link State Packet (LSP), or send the allocation result to each RB in the group through an intra-portal link between members of a multi-device link aggregation group (MCLAG);
the notification unit is further configured to send, in the peer-to-peer network, the allocation result to each RB in the group through an ESADI extension, or flood to the TRILL network through a Link State Packet (LSP), or send the allocation result to each RB in the group on an intra-portal link between MCLAG members.
Accordingly, in this embodiment, a routing bridge, applied in a TRILL network, includes:
a receiving unit, configured to receive a message sent from a remote end or a message sent from another Routing Bridge (RB) in a routing bridge group where the remote end is located;
a judging unit, configured to judge whether the message is a message that cannot be processed by the receiving unit and has a destination address of the routing bridge group after the receiving unit receives the message sent by the remote end; the routing bridge is also used for judging whether a message from a far end, which is received by the receiving unit and sent by other RBs, is a response message which is sent by the receiving unit and corresponds to a message packaged by using the Nickname of the routing bridge group when the routing bridge is used as a group designated node (GDRB) in the routing bridge group where the routing bridge is located;
the sending unit is used for sending the message to the GDRB in the routing bridge group after the judging unit judges that the message sent from the remote end is the message which can not be processed by the sending unit and the destination address of the message is the routing bridge group; and when the judging unit judges that the message from the remote end sent by the other RB received by the receiving unit is not the response message corresponding to the message encapsulated by the Nickname of the routing bridge group sent by the judging unit when the routing bridge is used as the GDRB in the routing bridge group where the routing bridge is located, the judging unit is further configured to respectively forward the message to other member RBs except the member RB which sent the message.
Preferably, the first and second liquid crystal films are made of a polymer,
and the sending unit is further configured to send a GDRB identity packet indicating that the routing bridge is a GDRB to other RBs in the group when determining that the routing bridge is configured as the GDRB.
Preferably, the first and second liquid crystal films are made of a polymer,
and the election unit is used for electing one RB from all RBs in the routing bridge group where the routing bridge is located as the GDRB according to an election rule before the GDRB identification message is not received.
Preferably, the first and second liquid crystal films are made of a polymer,
the election unit elects the GDRB according to the following election rules:
in the link aggregation group network, the election rule is that the RB with the highest priority of the root of the tree is preferentially elected in the group as the GDRB; or,
in the local area network, the election rule is that the RB with the highest priority of the interface in the group is preferentially elected as the GDRB; or,
in the region, the election rule is that one RB is selected as a GDRB according to the Nickname of each RB in the group.
It will be understood by those skilled in the art that all or part of the steps of the above methods may be implemented by instructing the relevant hardware through a program, and the program may be stored in a computer readable storage medium, such as a read-only memory, a magnetic or optical disk, and the like. Alternatively, all or part of the steps of the above embodiments may be implemented using one or more integrated circuits. Accordingly, each module/unit in the above embodiments may be implemented in the form of hardware, and may also be implemented in the form of a software functional module. The present invention is not limited to any specific form of combination of hardware and software.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Various other embodiments may be devised in accordance with the teachings of the present invention without departing from the spirit and scope thereof, and it should be understood that various changes and modifications may be effected therein by those skilled in the art without departing from the spirit and scope of the invention.

Claims (24)

1. A method for tree root distribution, which is applied in a transparent interconnection of lots of links (TRILL) network, comprises the following steps:
in the routing bridge group, one Routing Bridge (RB) is used as a group designated node (GDRB), a tree root is distributed to each RB in the group according to a tree selection strategy, and then the distribution result is notified to each RB in the group;
when a part of RB in the routing bridge group detects that distribution tree information to be used is configured on the equipment, the RB sends the distribution tree information to a GDRB in the routing bridge group;
the allocating tree roots for each RB in the group according to the tree selection strategy specifically includes:
after receiving the configured distribution tree information sent by other RBs in the group, the GDRB correspondingly preferentially allocates the tree to the RB; for each RB which is not configured with distribution tree information in the group, the GDRB selects a tree root with the shortest link cost or the largest link bandwidth with the RB from all currently unallocated tree roots to allocate to the RB.
2. The method of claim 1, wherein:
the GDRB in the routing bridge group is configured by the designation of an administrator, or each RB in the routing bridge group is elected from all RBs in the routing bridge group according to the same election rule.
3. The method of claim 2, wherein:
under the condition that the GDRB is configured by the specification of an administrator, when the GDRB judges that the GDRB is configured as the GDRB, the GDRB sends GDRB identification messages for indicating that the GDRB is the GDRB to other RBs in the group;
and after receiving the GDRB identification message, the RB in the group is known to be selected as the GDRB.
4. The method of claim 3, wherein:
the GDRB is elected from all RBs in the routing bridge group according to the same election rule by each RB in the routing bridge group, and specifically includes:
before the GDRB identification message is not received, selecting one RB from all RBs in the routing bridge group as the GDRB according to the same selection rule.
5. The method of claim 2 or 4, wherein:
in the link aggregation group network, the election rule is that the RB with the highest priority of the root of the tree is preferentially elected in the group as the GDRB; or,
in the local area network, the election rule is that the RB with the highest priority of the interface in the group is preferentially elected as the GDRB; or,
in the region, the election rule is that one RB is selected as a GDRB according to the Nickname of each RB in the group.
6. The method of claim 1, wherein:
the allocating tree roots for each RB in the group according to the tree selection strategy specifically includes:
the GDRB sorts all RBs in the group, and then tree roots are sequentially distributed to all the RBs in the group according to the sorting sequence; for each RB in the group, the GDRB selects a tree root with the shortest link cost or the largest link bandwidth from all currently unallocated tree roots to allocate to the RB.
7. The method of claim 1, wherein:
after receiving the configured distribution tree information sent by other RBs in the group, the GDRB correspondingly preferentially allocates the tree to the RB, specifically including:
after receiving the configured distribution tree information sent by other RBs in the group, the GDRB preferentially allocates the preconfigured distribution tree to an RB with a high tree root priority for two or more RBs with the same preconfigured distribution tree information, and selects a tree root with the shortest link cost or the largest link bandwidth from all currently unallocated tree roots to allocate to the other RBs.
8. The method of claim 1, wherein:
the notifying of the allocation result to each RB in the group specifically includes:
in a local area network, sending the distribution result to each RB in a group through a HELLO message in a TRILL network, or sending the distribution result to each RB in the group through a terminal address interaction protocol (ESADI) extension, or flooding to the TRILL network through a link state data packet (LSP), or sending the distribution result to each RB in the group through an intra-portal link among members of a multi-device link aggregation group (MCLAG);
and in the point-to-point network, the distribution result is sent to each RB in the group through ESADI extension, or is flooded to a TRILL network through a link state data packet (LSP), or is sent to each RB in the group on an intra-portal link among MCLAG members.
9. A message processing method is applied to a transparent interconnection of lots of links (TRILL) network, and comprises the following steps:
in the routing bridge group, one Routing Bridge (RB) is used as a group designation node (GDRB);
after a member Routing Bridge (RB) receives a message sent from a remote end, if the message is judged to be a message which can not be processed by the member Routing Bridge (RB) and the destination address of the message is the message of the routing bridge group, the message is sent to a GDRB in the group;
after receiving the message, the GDRB respectively forwards the message to other member RBs except the member RB from which the message is sent if judging that the message is not a response message corresponding to the message which is sent by the GDRB and packaged by using the Nickname of the routing bridge group;
after receiving the configured distribution tree information sent by other RBs in the group, the GDRB correspondingly preferentially allocates the tree to the RB; for each RB which is not configured with distribution tree information in the group, the GDRB selects a tree root with the shortest link cost or the largest link bandwidth with the RB from all currently unallocated tree roots to allocate to the RB;
when a part of RB in the routing bridge group detects that the distribution tree information to be used is configured on the device, the RB sends the distribution tree information to the GDRB in the routing bridge group.
10. The method of claim 9, wherein:
the GDRB in the routing bridge group is configured by the designation of an administrator, or each RB in the routing bridge group is elected from all RBs in the routing bridge group according to the same election rule.
11. The method of claim 10, wherein:
under the condition that the GDRB is configured by the specification of an administrator, when the GDRB judges that the GDRB is configured as the GDRB, the GDRB sends GDRB identification messages for indicating that the GDRB is the GDRB to other RBs in the group;
and after receiving the GDRB identification message, the RB in the group is known to be selected as the GDRB.
12. The method of claim 11, wherein:
the GDRB is elected from all RBs in the routing bridge group according to the same election rule by each RB in the routing bridge group, and specifically includes:
before the GDRB identification message is not received, selecting one RB from all RBs in the routing bridge group as the GDRB according to the same selection rule.
13. The method of claim 10 or 12, wherein:
in the link aggregation group network, the election rule is that the RB with the highest priority of the root of the tree is preferentially elected in the group as the GDRB; or,
in the local area network, the election rule is that the RB with the highest priority of the interface in the group is preferentially elected as the GDRB; or,
in the region, the election rule is that one RB is selected as a GDRB according to the Nickname of each RB in the group.
14. A routing bridge for use in a transparent interconnection of lots of links (TRILL) network, comprising:
a tree selection unit, configured to allocate a tree root to each Routing Bridge (RB) in a group according to a tree selection policy when the RB serves as a group designation node (GDRB) in a routing bridge group;
a notification unit, configured to notify the distribution result of the tree root distributed by the tree selection unit for each RB to each RB in the group;
the notification unit is further configured to send distribution tree information to be used to the GDRB in the routing bridge group where the routing bridge is located when detecting that the device has configured the distribution tree information to be used;
the tree selecting unit is configured to allocate tree roots for each RB in the group according to a tree selecting policy, and specifically includes:
the tree selecting unit is used for correspondingly and preferentially allocating the tree to the RB after receiving the configured distribution tree information sent by other RBs in the group; and for each RB which is not configured with the distribution tree information in the group, selecting the root with the shortest link cost or the largest link bandwidth from all the currently unallocated roots to allocate to the RB.
15. The routing bridge of claim 14, wherein:
and the notification unit is further configured to send a GDRB identity packet indicating that the routing bridge is a GDRB to other RBs in the group when determining that the routing bridge is configured as the GDRB.
16. The routing bridge of claim 15, further comprising:
and the election unit is used for electing one RB from all RBs in the routing bridge group where the routing bridge is located as the GDRB according to an election rule before the GDRB identification message is not received.
17. The routing bridge of claim 16, wherein:
the election unit elects the GDRB according to the following election rules:
in the link aggregation group network, the election rule is that the RB with the highest priority of the root of the tree is preferentially elected in the group as the GDRB; or,
in the local area network, the election rule is that the RB with the highest priority of the interface in the group is preferentially elected as the GDRB; or,
in the region, the election rule is that one RB is selected as a GDRB according to the Nickname of each RB in the group.
18. The routing bridge of claim 14, wherein:
the tree selecting unit is configured to allocate tree roots for each RB in the group according to a tree selecting policy, and specifically includes:
the tree selecting unit is used for sequencing all RBs in the group and then sequentially distributing tree roots for the RBs in the group according to the sequencing sequence; and the tree selecting unit is used for selecting a tree root with the shortest link cost or the largest link bandwidth from all currently unallocated tree roots for each RB in the group to allocate to the RB.
19. The routing bridge of claim 14, wherein:
the tree selection unit is configured to, after receiving the configured distribution tree information sent by other RBs in the group, correspondingly preferentially allocate the tree to the RB, and specifically includes:
and the tree selecting unit is used for preferentially allocating the pre-configured distribution tree to the RB with the higher tree root priority for more than two RBs with the same pre-configured distribution tree information after receiving the configured distribution tree information sent by other RBs in the group, and selecting the tree root with the shortest link cost or the largest link bandwidth from all the currently unallocated tree roots to allocate to other RBs.
20. The routing bridge of claim 14, wherein:
the notifying unit is configured to notify each RB in the group of the allocation result, and specifically includes:
the notification unit is configured to send, in the local area network, the allocation result to each RB in the group through a HELLO message in the TRILL network, or send the allocation result to each RB in the group through an end address interaction protocol (ESADI) extension, or flood the allocation result to the TRILL network through a Link State Packet (LSP), or send the allocation result to each RB in the group through an intra-portal link between members of a multi-device link aggregation group (MCLAG);
the notification unit is further configured to send, in the peer-to-peer network, the allocation result to each RB in the group through an ESADI extension, or flood to the TRILL network through a Link State Packet (LSP), or send the allocation result to each RB in the group on an intra-portal link between MCLAG members.
21. A routing bridge for use in a transparent interconnection of lots of links (TRILL) network, comprising:
a receiving unit, configured to receive a message sent from a remote end or a message sent from another Routing Bridge (RB) in a routing bridge group where the remote end is located;
a judging unit, configured to judge whether the message is a message that cannot be processed by the receiving unit and has a destination address of the routing bridge group after the receiving unit receives the message sent by the remote end; the routing bridge is also used for judging whether a message from a far end, which is received by the receiving unit and sent by other RBs, is a response message which is sent by the receiving unit and corresponds to a message packaged by using the Nickname of the routing bridge group when the routing bridge is used as a group designated node (GDRB) in the routing bridge group where the routing bridge is located; the routing bridge group is also used for sending the distribution tree information to the GDRB in the routing bridge group when part of the RB in the routing bridge group detects that the distribution tree information to be used is configured on the equipment;
the sending unit is used for sending the message to the GDRB in the routing bridge group after the judging unit judges that the message sent from the remote end is the message which can not be processed by the sending unit and the destination address of the message is the routing bridge group; the routing bridge is also used for respectively forwarding the message to other member RBs except the member RB which sends the message when the judging unit judges that the message from the far end sent by other RBs received by the receiving unit is not the response message which is sent by the judging unit and corresponds to the message which is packaged by using the Nickname of the routing bridge group and sent by the judging unit when the routing bridge is used as the GDRB in the routing bridge group where the routing bridge is located; after receiving the configured distribution tree information sent by other RBs in the group, the system is used for correspondingly preferentially distributing the tree to the RB; for each RB which is not configured with distribution tree information in the group, the GDRB selects a tree root with the shortest link cost or the largest link bandwidth with the RB from all currently unallocated tree roots to allocate to the RB.
22. The routing bridge of claim 21, wherein:
and the sending unit is further configured to send a GDRB identity packet indicating that the routing bridge is a GDRB to other RBs in the group when determining that the routing bridge is configured as the GDRB.
23. The routing bridge of claim 22, further comprising:
and the election unit is used for electing one RB from all RBs in the routing bridge group where the routing bridge is located as the GDRB according to an election rule before the GDRB identification message is not received.
24. The routing bridge of claim 23, wherein:
the election unit elects the GDRB according to the following election rules:
in the link aggregation group network, the election rule is that the RB with the highest priority of the root of the tree is preferentially elected in the group as the GDRB; or,
in the local area network, the election rule is that the RB with the highest priority of the interface in the group is preferentially elected as the GDRB; or,
in the region, the election rule is that one RB is selected as a GDRB according to the Nickname of each RB in the group.
CN201210148395.0A 2012-05-14 2012-05-14 A kind of distribution of tree root, the method for Message processing and route-bridge Expired - Fee Related CN103428088B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210148395.0A CN103428088B (en) 2012-05-14 2012-05-14 A kind of distribution of tree root, the method for Message processing and route-bridge

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210148395.0A CN103428088B (en) 2012-05-14 2012-05-14 A kind of distribution of tree root, the method for Message processing and route-bridge

Publications (2)

Publication Number Publication Date
CN103428088A CN103428088A (en) 2013-12-04
CN103428088B true CN103428088B (en) 2018-11-06

Family

ID=49652278

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210148395.0A Expired - Fee Related CN103428088B (en) 2012-05-14 2012-05-14 A kind of distribution of tree root, the method for Message processing and route-bridge

Country Status (1)

Country Link
CN (1) CN103428088B (en)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104702506B (en) * 2013-12-09 2018-10-19 华为技术有限公司 A kind of message transmitting method, network node and message transfer system
EP3694158B1 (en) * 2014-05-13 2022-01-05 Huawei Technologies Co., Ltd. Active-active access to transparent interconnection of lots of links (trill) edges
CN105306613A (en) 2014-07-24 2016-02-03 中兴通讯股份有限公司 MAC address notification method and device and acquisition device for ESADI
CN104243321A (en) * 2014-09-05 2014-12-24 杭州华三通信技术有限公司 Selection method and routing bridge (RB) devices for distribution tree root
CN105591940B (en) * 2014-11-17 2020-03-31 中兴通讯股份有限公司 TRILL network distribution tree selection method and TRILL network node
CN105681204A (en) * 2014-11-21 2016-06-15 中兴通讯股份有限公司 Transparent interconnection of lots of links (TRILL) network distribution tree calculation method and TRILL network node
CN105704257A (en) * 2014-11-28 2016-06-22 中兴通讯股份有限公司 Media access control (MAC) address election method and device
CN106941446A (en) * 2016-01-04 2017-07-11 中兴通讯股份有限公司 Locally aggregated chain processing method and device
CN108989208B (en) * 2017-05-31 2021-08-10 中兴通讯股份有限公司 Identification distribution method and device
CN109388490B (en) * 2017-08-07 2020-11-17 华为技术有限公司 Memory allocation method and server
CN109688060B (en) * 2018-12-29 2021-06-29 杭州迪普科技股份有限公司 Link grouping configuration method and device and router

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101136843A (en) * 2006-08-30 2008-03-05 华为技术有限公司 Method of generating symmetrical tree in the shortest path bridge
CN102299845A (en) * 2011-09-20 2011-12-28 神州数码网络(北京)有限公司 Multi-link transparent interconnection multicast frame transmission method and system thereof
CN102316033A (en) * 2011-09-06 2012-01-11 中兴通讯股份有限公司 Method for assigning distribution trees and system thereof

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101136843A (en) * 2006-08-30 2008-03-05 华为技术有限公司 Method of generating symmetrical tree in the shortest path bridge
CN102316033A (en) * 2011-09-06 2012-01-11 中兴通讯股份有限公司 Method for assigning distribution trees and system thereof
CN102299845A (en) * 2011-09-20 2011-12-28 神州数码网络(北京)有限公司 Multi-link transparent interconnection multicast frame transmission method and system thereof

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
Routing Bridges:Base Protocol Specification;RFC6325;《IETF》;20110731;第52-55页 *

Also Published As

Publication number Publication date
CN103428088A (en) 2013-12-04

Similar Documents

Publication Publication Date Title
CN103428088B (en) A kind of distribution of tree root, the method for Message processing and route-bridge
EP3949293B1 (en) Slice-based routing
EP3624408B1 (en) Method for generating forwarding table entry, controller, and network device
US11689452B2 (en) Method for forwarding service data, network device, and network system
CN102150148B (en) Differentiated services for unicast multicast frames in layer 2 topologies
US8995444B2 (en) Method and system for extending routing domain to non-routing end stations
CN112087386B (en) Message processing method, device and system
US8009684B2 (en) High capacity ring communication network
EP3934183B1 (en) Service function chain sfc-based communication methods, and apparatuses
CN102055665B (en) OSPF point-to-multipoint over broadcast or NBMA mode
CN102415065A (en) Redundant host connection in a routed network
EP3883286A1 (en) Data processing method, controller and forwarding device
US9641433B2 (en) Method, routing bridge, and system for sending packet
WO2021164402A1 (en) Routing method, routing device and computer-readable storage medium
WO2022166465A1 (en) Message processing method and related apparatus
WO2023274083A1 (en) Route publishing method and apparatus, packet forwarding method and apparatus, device, and storage medium
CN112822097A (en) Message forwarding method, first network device and first device group
WO2022048418A1 (en) Method, device and system for forwarding message
WO2022007702A1 (en) Message processing method and network device
CN112751766A (en) Message forwarding method and device and computer storage medium
CN114760244B (en) Method, device and network equipment for transmitting Binding Segment Identification (BSID)
CN112737946B (en) Route advertising method, device, storage medium and system for IPv6 network
CN115811499A (en) Method and equipment for designated forwarder DF election
CN113595915A (en) Method for forwarding message and related equipment
CN108259292B (en) Method and device for establishing tunnel

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20181106

Termination date: 20200514