CN118337697A - Method and device for sending routing message and cross-layer data message - Google Patents

Abstract

The embodiment of the invention provides a method and a device for sending a routing message and a cross-layer data message, wherein a first message from a second node is received through a first node, and the first message comprises the second message; the first node judges whether a node corresponding to the third indication information meets a first condition, if yes, the first processing is executed, if not, the second processing is executed, wherein the first processing is to prohibit sending of the second message to the node corresponding to the third indication information, the second processing is to send the second message to the node corresponding to the third indication information, the first condition is that the first node receives a third route issued by the second node, and the third route is a route carrying the third indication information. The problem that the limit that the BUM message received from vHub cannot be forwarded to the vSpoke node cannot be released in the related art is solved, and the effect that the limit is released and a loop is not caused is achieved.

Description

Routing message, and cross-layer data message sending method and device

Technical Field

The embodiment of the invention relates to the field of communication, in particular to a method and a device for transmitting cross-layer data messages and routing messages.

Background

Ethernet virtual private network (Ethernet Virtual Private Network, EVPN) is a technology based on border gateway protocol (Border Gateway Protocol, BGP) and multiprotocol label switching (Multi-Protocol Label Switching, MPLS) protocols, and is widely used in private network operation, data center, etc. Headquarters and various branches located in different regions are connected to Provider Edge (PE) devices through respective Customer Edge (CE) devices, where, as shown in fig. 1, PE devices (denoted as VH1 and VH 2) for the headquarter CE devices to access are PE devices of Virtual Hub type, which may be referred to as Virtual Hub (vHub) nodes or Hub nodes, and so on; PE devices (denoted as VS1, VS2, VS3, respectively) for access by the fractional CE devices (not shown in fig. 1) are PE devices of the Virtual Spoke type, which may be referred to as Virtual Spoke (vSpoke) nodes or Spoke nodes, etc. In Hub/Spoke networking, spoke nodes and Hub nodes can communicate with each other, but direct communication is not allowed between Spoke nodes, so that communication between Spoke nodes needs to be achieved by using Hub nodes, for example, when VS1 needs to broadcast a message to VS2, then the message needs to be sent to VH1 (or VH 2) first and then forwarded to VS2 by VH1 (or VH 2).

Broadcast, unknown unicast and multicast (Broadcast Unicast Multicast, BUM) messages transmitted between vHub nodes in the same cluster are defined in the conventional technology, and a vHub node serving as a receiving end cannot forward the BUM message to other vSpoke nodes in the cluster any more so as to avoid loops. For example, VH1 cannot forward BUM messages received from VH2 to any of VS1/VS2/VS 3.

Disclosure of Invention

The embodiment of the invention provides a routing message and a cross-layer data message sending method and device, which at least solve the problem that the limit that a BUM message received from vHub cannot be forwarded to a vSpoke node cannot be released in the related technology.

According to one embodiment of the present invention, there is provided a cross-layer data message sending method, including: the method comprises the steps that a first node receives a first message from a second node, wherein the first message comprises a second message; the first node judges whether a node corresponding to third indication information meets a first condition, if yes, first processing is executed, if not, second processing is executed, wherein the first processing is to prohibit sending of the second message to the node corresponding to the third indication information, the second processing is to send the second message to the node corresponding to the third indication information, the first condition is that the first node receives a third route issued by the second node, and the third route is a route carrying the third indication information.

According to another embodiment of the present invention, there is provided a routing message sending method, including: the second node receives the first route; the second node sends a third route to the first node, wherein the third route carries third indication information, the third indication information is used for indicating the first node to execute at least one of first processing and third processing in the third route, the first processing is a node which prohibits forwarding of a second message to the first route, the node corresponds to the third indication information, the third processing is a node which performs DF election according to the third route, and decides whether to forward a fifth message to the node which issues the first route according to an election result of the DF election, the fifth message is from a second layer node, the second layer node is a node outside a first cluster, and the first cluster is a cluster where the first node, the second node and the node which issues the first route are located. According to another embodiment of the present invention, there is provided a method for retransmitting a routing packet, including: and receiving the automatic discovery route of each ES, and when an eighth condition is met, modifying the horizontal division label in the automatic discovery route of each ES, and then releasing the automatic discovery route of each ES, wherein the eighth condition is that a route target carried in the automatic discovery route of each ES is matched with a leading-in route target of a local MAC-VRF.

According to another embodiment of the present invention, there is provided a cross-layer data packet transmission apparatus, including: the first receiving module is used for receiving a first message from a second node, wherein the first message comprises a second message; the first sending module is configured to determine whether a node corresponding to the third indication information meets a first condition, execute a first process if the node meets the first condition, and execute a second process if the node does not meet the first condition, where the first process is to prohibit sending the second message to the node corresponding to the third indication information, and the second process is to send the second message to the node corresponding to the third indication information, where the first condition is that the first node receives a third route issued by the second node, and the third route is a route carrying the third indication information.

According to still another embodiment of the present invention, there is also provided a routing packet transmitting apparatus, including: a second receiving module for receiving the first route; the second sending module is configured to send a third route to the first node, where the third route carries third indication information, where the third indication information is used to instruct the first node to perform at least one of a first process and a third process in the third route, where the first process is a node that prohibits forwarding of a second packet to the first route, which corresponds to the third indication information, and the third process is that the second node performs DF election according to the third route, and decides whether to forward a fifth packet to the first route, according to an election result of the DF election specific to the first route, where the fifth packet is from a second layer node, where the second layer node is a node outside a first cluster, and where the first node, the second node, and the first route are located.

According to an embodiment of the present invention, there is provided a routing packet retransmission apparatus, including: and the retransmission module is used for receiving the automatic discovery route of each ES, modifying the horizontal segmentation label in the automatic discovery route of each ES when an eighth condition is met, and then releasing the automatic discovery route of each ES, wherein the eighth condition is that a route target carried in the automatic discovery route of each ES is matched with a local MAC-VRF import route target.

According to a further embodiment of the invention, there is also provided a computer readable storage medium having stored therein a computer program, wherein the computer program is arranged to perform the steps of any of the method embodiments described above when run.

According to a further embodiment of the invention, there is also provided an electronic device comprising a memory having stored therein a computer program and a processor arranged to run the computer program to perform the steps of any of the method embodiments described above.

The invention provides a cross-layer data message sending method, which comprises the steps of receiving a first message from a second node through a first node, wherein the first message comprises the second message; the first node judges whether a node corresponding to the third indication information meets a first condition, if yes, the first processing is executed, if not, the second processing is executed, wherein the first processing is to prohibit sending of the second message to the node corresponding to the third indication information, the second processing is to send the second message to the node corresponding to the third indication information, the first condition is that the first node receives a third route issued by the second node, and the third route is a route carrying the third indication information. The problem that a limit that a BUM message received from vHub cannot be forwarded to a vSpoke node is solved, and a loop is caused is solved, so that the effect that the message can be transmitted and the loop is not caused is achieved.

Drawings

FIG. 1 is a schematic diagram of a Hub/Spoke networking framework employing EVPN technology;

FIG. 2 is a schematic diagram of a Hub/Spoke per cluster networking framework employing EVPN technology;

fig. 3 is a schematic diagram of a forwarding trace of a BUM message in an EVPN AR networking;

fig. 4 is a flowchart of a cross-layer data packet sending method provided by an embodiment of the present invention;

FIG. 5 is a flowchart of a cross-layer data message sending method provided by an embodiment of the present invention;

fig. 6 is a flowchart of a method of routing a message according to an embodiment of the present invention;

FIG. 7 is a flow chart of a routing message resending method according to an embodiment of the present invention;

fig. 8 is a block diagram of a cross-layer data packet transmission apparatus according to an embodiment of the present invention;

fig. 9 is a block diagram of a routing message transmitting apparatus according to an embodiment of the present invention;

fig. 10 is a block diagram of a routing message retransmission apparatus according to an embodiment of the present invention;

FIG. 11 is a forwarding trace diagram of a second message of a cross-layer data messaging method according to an embodiment of the present invention;

Fig. 12 is a flow chart of a data message forwarding principle according to an embodiment of the present invention;

FIG. 13 is a flow chart of a routing message sending method according to an embodiment of the present invention;

FIG. 14 is a forwarding trace diagram of a second message of a cross-layer data messaging method according to an embodiment of the present invention;

fig. 15 is a schematic diagram of a routing message forwarding flow according to an embodiment of the present invention;

Fig. 16 is a schematic diagram of a routing message forwarding flow in accordance with an embodiment of the present invention;

fig. 17 is a schematic diagram of a routing message forwarding flow according to an embodiment of the present invention;

FIG. 18 is a diagram illustrating the location of data message reception/transmission in accordance with an embodiment of the present invention;

FIG. 19 is a schematic diagram of an auto discovery route redistribution per ES Ethernet in accordance with an embodiment of the present invention;

Fig. 20 is a schematic diagram of an auto discovery route redistribution per ES ethernet in accordance with an embodiment of the present invention.

Detailed Description

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings in conjunction with the embodiments.

It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order.

First, to further analyze the problems in the prior art, fig. 1 is a schematic diagram of a Hub/Spoke networking framework using an EVPN technology, and as shown in fig. 1, an exemplary application scenario is deployed using the EVPN technology "draft-ietf-bess-EVPN-virtual-Hub-00", where the scenario may be Hub/Spoke networking of a certain organization or company. However, communication between Spoke nodes needs to be achieved by Hub nodes, for example, when VS1 needs to broadcast a message to VS2, the message needs to be sent to VH1 (or VH 2) and then forwarded to VS2 by VH1 (or VH 2).

The concept of clusters (Cluster) is also defined in "draft-ietf-bess-EVPN-virtual-Hub-00", FIG. 2 is a schematic diagram of Hub/Spoke by Cluster networking framework using EVPN technology, as shown in FIG. 2, the enterprise has two clusters, cluster1 consists of VH1, VH2, VS1, VS2, VS3, cluster2 consists of VH3, VH4, VS5, VS6, wherein VH1, VH2, VH3, VH4 are vHub nodes, and the others are vSpoke nodes. When one vHub receives a BUM message from another vHub in the present cluster, it cannot be forwarded to any one vSpoke node in the present cluster. For any one of the clusters, any one node in the other cluster is a node outside the cluster.

The EVPN-assisted replication AR networking is another hierarchical networking defined by "draft-ietf-bess-EVPN-optimized-ir," and the EVPN AR networking differs from Hub/Spoke networking in terms of known unicast message forwarding and signaling layers, but BUM message forwarding in the EVPN AR networking is similar to BUM message forwarding in the Hub/Spoke networking. Fig. 3 is a schematic diagram of a forwarding trace of a BUM packet in an EVPN AR networking, where ar_r2 and ar_r1 are AR replicator nodes, and ar_l1, ar_l2 and ar_l3 are AR leaf nodes, and for a BUM packet forwarded from ar_r2 to ar_r1 in an ingress replication mode, ar_r1 cannot be forwarded to an ar_l1, ar_l2 or ar_l3 node any more, so that when any node of ar_r2 to ar_l1, ar_l2 or ar_l3 is not reachable and ar_r1 to that node (denoted as T node) is reachable, then a node redundancy group consisting of ar_r1 and ar_r2 nodes will result in a BUM packet that should be received by that node (such as T node) not reaching that node. It should be noted that, if ar_r1/ar_r2 in fig. 3 is replaced by vHub nodes (respectively called VH1/VH 2) in Hub/Spoke networking, and each AR leaf node (ar_l1, ar_l2, and ar_l3) in fig. 3 is replaced by vSpoke nodes (respectively called VS1, VS2, and VS 3) in Hub/Spoke networking, the above problem is the same for the BUM message forwarding flow in the network after such modification, that is: for BUM messages forwarded from ar_r2 to ar_r1 in the ingress copy mode, ar_r1 cannot be forwarded to ar_l1, ar_l2 or ar_l3 nodes anymore, and therefore when any one of ar_r2 to ar_l1, ar_l2 or ar_l3 nodes is not reachable and ar_r1 to that node (denoted as T node) is reachable, then the redundant set of nodes consisting of ar_r1 and ar_r2 nodes will cause BUM packets that should be received by that node (such as T node) to not reach that node.

Once the vHub node (or AR replicator node) can forward the BUM message received from the other vHub node (or AR replicator node) to the vSpoke node (or AR leaf node), not only can the BUM message received in the second step be sent to the ar_l2 in the fourth step, but also the ar_r1 can send the BUM message to the ar_l3 in the fifth step, because the BUM message comes from the ar_l3, a loop is formed, and thus the limitation that the BUM message received from the vHub node (or AR replicator node) must not be forwarded to the vSpoke node (or AR leaf node) cannot be released in the prior art, which causes a loop.

It should be noted that, for convenience of description, in the present invention, forwarding a copy of a given message may sometimes be directly referred to as forwarding the message, and receiving a copy of a given message may sometimes be directly referred to as receiving the message.

In this embodiment, a method for sending a cross-layer data packet is provided, and fig. 4 is a flowchart of the method for sending a cross-layer data packet provided in the embodiment of the present invention, as shown in fig. 4, where the flowchart includes:

Step S402, a first node receives a first message from a second node, wherein the first message comprises a second message;

Step S404, the first node judges whether the node corresponding to the third indication information meets a first condition, if yes, the first processing is executed, and if not, the second processing is executed, wherein the first processing is to prohibit sending the second message to the node corresponding to the third indication information, the second processing is to send the second message to the node corresponding to the third indication information, wherein the first condition is that the first node receives a third route issued by the second node, and the third route is a route carrying the third indication information.

It should be noted that, in some embodiments, other nodes may also be included in the first layer node.

Before step S404, for each node other than the first layer node in the first cluster, the first node regards the node as a node corresponding to the corresponding third indication information in the third layer node, and executes step S404, where the sending of the second message to the node corresponding to the corresponding third indication information in the third layer node in step S404 may be sending different copies of the second message to the node corresponding to the corresponding third indication information in the third layer node.

In an exemplary embodiment, the third indication information is indication information corresponding to a node in a third layer of nodes, the third layer of nodes is formed by nodes other than the first layer of nodes in the first cluster, the first cluster is a cluster in which the first node and the second node are located, and the first layer of nodes is a node in which the first node and the second node are located.

In one exemplary embodiment, before the second process is performed, further comprising: and deciding whether to execute the second process according to DF election results specific to the node corresponding to the third indication information.

In an exemplary embodiment, before the first node receives the first message from the second node, the method further comprises: the first node issues a first label to the second node, wherein the first message comprises the first label, and the first label is the label from which the first node determines that the first message comes from the second node. Fig. 5 is a flowchart of a cross-layer data packet sending method according to an embodiment of the present invention, where, as shown in fig. 5, the flowchart includes:

step S502, a first node issues a first label to a second node, wherein the first label comprises the first label, and the first label is the label from which the first node determines that the first message comes from the second node;

step S504, the first node receives a first message from a second node, wherein the first message comprises a second message;

In step S506, the first node determines whether the node corresponding to the third indication information meets a first condition, if yes, the first process is executed, and if not, the second process is executed, wherein the first process is to prohibit sending the second message to the node corresponding to the third indication information, the second process is to send the second message to the node corresponding to the third indication information, wherein the first condition is that the first node receives a third route issued by the second node, and the third route is a route carrying the third indication information.

In an exemplary embodiment, the first packet further includes: a first horizontal split tag, wherein the first horizontal split tag corresponds to a first ethernet segment identifier (ETHERNET SEGMENT IDENTIFIER, ESI), the first ESI identifying a first ethernet segment (ETHERNET SEGMENT, ES), the first ES not being contiguous with the first node and the second node; the first node adds a third horizontal segmentation tag to the second message to obtain a third message, the first node sends the third message to a third layer node, the third horizontal segmentation tag is used for indicating the third layer node to prohibit forwarding of the second message to the first ES, the value of the third horizontal segmentation tag is equal to the value of the first horizontal segmentation tag or the value of the third horizontal segmentation tag meets a third condition, and the third condition is that the value of the third horizontal segmentation tag is determined according to the value of the first horizontal segmentation tag.

It should be noted that, in an exemplary embodiment, the ethernet auto-discovery route of each ES may be an ethernet auto-discovery route with ESI of 0, the horizontal split label may be a Leaf label carried in an E-Tree extended group attribute, and the first horizontal split label corresponds to the first ESI and may be determined by an access circuit AC on the first ES being an AC of the E-Tree Leaf attribute.

In an exemplary embodiment, the first node sends the second message to the third layer node, including: the first message further comprises a virtual extended local area network (Virtual eXtensible Local Area Network, VXLAN) encapsulation, the first node obtains a first horizontal split label from a user datagram protocol (User Datagram Protocol, UDP) source port of the VXLAN encapsulation of the first message, the first node performs VXLAN encapsulation on the second message to obtain a third message, wherein the source IP of the tunnel of the VXLAN is the first node, and the UDP source port of the VXLAN carries the third horizontal split label.

The embodiment of the invention also provides a routing message sending method, and fig. 6 is a flowchart of the routing message sending method according to the embodiment of the invention, as shown in fig. 6, and the flowchart includes:

step S602, a second node receives a first route;

In step S604, the second node sends a third route to the first node, where the third route carries third indication information, where the third indication information is used to instruct the first node to perform at least one of a first process and a third process in the third route, where the first process is to prohibit forwarding of the second message to a node that issues the first route corresponding to the third indication information, the third process is to perform designated forwarder (DESIGNATED FORWARDER, DF) election according to the third route by the second node, and determine whether to forward the fifth message to the node that issues the first route according to an election result of the DF election specific to the node that issues the first route, where the fifth message is from a second layer node, where the second layer node is a node other than the first cluster, where the first node, the second node, and the node that issues the first route are located.

In one exemplary embodiment, further comprising: the node for issuing the first route is a node in a third layer of nodes in the first cluster, the third layer of nodes consists of nodes except the first layer of nodes in the first cluster, and the first layer of nodes are nodes of a layer where the first node and the second node are located;

The second node receives the second message, and the second node sends the second message to a fourth node in the third layer node under the condition that the second message received by the second node comes from the third node in the third layer node.

In one exemplary embodiment, the second node performs DF elections according to at least the third route, and the second node is a non-DF node specific to the node that issued the first route, as determined from the DF elections.

In one exemplary embodiment, further comprising: the second node receives a sixth message from the second layer node, and the second node prohibits sending the sixth message to the node issuing the first route.

It should be noted that, in some embodiments, the nodes in the first cluster include a first layer node and a third layer node, and the principle of dividing the first layer node and the third layer node is related to a specific scenario, for example: in one scenario, the first layer node is a Virtual Hub node, and the third layer node is a Virtual Spoke node; in another scenario, the first layer node is an AR-REPLICATOR node, and the third layer node is an AR-LEAF node; in another scenario, the first layer node is a node in the first cluster that directly interacts with a node outside the first cluster with a virtual private network (Virtual Private Network, VPN) route, and the third layer node is a node in the first cluster that interacts with the node outside the first cluster through the first layer node.

In an actual implementation process, the sixth message and the fifth message are messages with the same property, but the fifth message is received by the first node from outside the first cluster, and the sixth message is received by the second node from outside the first cluster. Because the first node and the second node can be interchanged, the processing logic of the first node and the second node are identical in practice, and the forwarding rule of the fifth message can be refined through describing the sixth message.

In one exemplary embodiment, the second node performs DF elections according to at least a third route, including: and the second node receives a sixth route which is issued by the first node and carries third indication information, and the first node performs DF election according to the third route and the sixth route.

In an exemplary embodiment, the third route and the sixth route comprise an operator edge device identification Label Attribute (PED Label(s) Attribute, PEDLA) carrying a PE identifier (PED) of the operator edge device PE, the PED identifier being used as the third indication information for identifying the node issuing the first route, the DF election determined DF node being specific to the PED identifier.

In one exemplary embodiment, the first route includes one of: a inclusive multicast ethernet label IMET route, wherein an initiator router internet address (Originating RouterIP Address, ORIP) in the inclusive multicast ethernet label (Inclusive Multicast ETHERNET TAG, IMET) route carries a third indication information; the ethernet auto-discovery route per EVPN instance (EVPN INSTANCE, EVI) wherein the third indication information is an indication information corresponding to a first EVPN VPWS signaling instance that is an EVPN VPWS signaling instance that matches the first ethernet label identifier ETI in the ethernet auto-discovery route per EVI.

It should be noted that the EVPN VPWS signaling instance may be part of the EVPN VPWS instance used for the signaling procedure in RFC 8214.

In a particular embodiment, the second node prohibits using the IMET route issued by the node issuing the first route for forwarding the BUM message in the event that the first route is not an IMET route (e.g., the first route is an auto-discovery route per EVI ethernet), and the first route is used for forwarding a broadcast, unknown unicast, or multicast BUM message.

In one exemplary embodiment, the first route includes third indication information identifying a node in the first route that issued the first route.

In an exemplary embodiment, the first EVPN VPWS signaling instance is determined by the first ethernet tag identifier ETI and the second ETI, and the third indication information is the second ETI or the third indication information is fourth indication information corresponding to the first EVPN VPWS signaling instance.

In a specific embodiment, the method further comprises: the second node binds a first bidirectional forwarding entity, BFE, to a MAC-VRF of the ethernet virtual private network, EVPN, the first BFE being a forwarding plane entity corresponding to the first EVPN VPWS signaling instance.

In one exemplary embodiment, the second node obtains the second horizontal split tag and the second message from the fourth message and determines the value of the first horizontal split tag according to the second horizontal split tag, encapsulates the second message and the first horizontal split tag in the first message, and sends the first message to the first node.

In one exemplary embodiment, the second node sends a first message to the first node, and the second node writes a tag issued by the first node in the first message that identifies the second node.

In one exemplary embodiment, the second node sends a third route to the first node, comprising: the third route includes a PED label attribute PEDLA, which carries a PE identifier PED corresponding to the node that issued the first route, and the second node sets PED label PEDL corresponding to PED in PEDLA to a first specified value.

It should be noted that, the first specified value may be a reserved tag value, such as an explicit null tag, an implicit null tag, or the like.

In the specific implementation process, the third node and the fourth node may be the same class of nodes, and in this case, different condition judgment is performed on the same class of nodes, so that different processes may be performed. Similarly, the fourth node may not be the same node as the third node, and in this case, the third node and the fourth node may still be subjected to the judgment of the same condition, and then different processes may be performed.

In one exemplary embodiment, where the first route is an auto-discover route per EVI ethernet, and where the first route is used to forward broadcast, unknown unicast, or multicast BUM messages, the second node prohibits use of IMET routes published by the node that published the first route for forwarding BUM messages.

The embodiment of the invention also provides a method for resending the routing message, and fig. 7 is a flow chart of the method for resending the routing message according to the embodiment of the invention, as shown in fig. 7, and comprises the following steps:

Step S702, receiving the automatic discovery route of each ES, modifying the horizontal split label in the automatic discovery route of each ES, and then issuing the automatic discovery route of each ES when the eighth condition is satisfied, wherein the eighth condition is that the route target carried in the automatic discovery route of each ES is matched with the import route target of the local MAC virtual forwarding instance (MAC Virtual Routing and Forwarding, MAC-VRF).

In one exemplary embodiment, further comprising: when the eighth condition is satisfied, one of the following processes is performed for each ES ethernet auto-discovery route:

deleting an initiator provider edge (Originating Provider Edge, OPE) type length Value (TYPE LENGTH Value, TLV) carried in the per ES ethernet auto-discovery route;

Writing first indication information into each ES Ethernet automatic discovery route, wherein the first indication information indicates to ignore an OPE TLV carried in each ES Ethernet automatic discovery route;

Covering the initiator PE identification information in the OPE TLV carried in the automatic discovery route of each ES by using local node identification information;

Writing second indication information and a seventh identifier into each ES ethernet automatic discovery route, where the second indication information indicates that, when the receiving end node of each ES ethernet automatic discovery route forwards a first type packet according to the first type route, a horizontal split label in each ES ethernet automatic discovery route is added, the first type route is a containing provider multicast service interface I-PMSI (Inclusive-Provider Multicast SERVICE INTERFACE) route with ORIP being the seventh identifier or a Selective provider multicast service interface S-PMSI (Selective-Provider Multicast SERVICE INTERFACE) route, and when the ESI of each ES ethernet automatic discovery route is 0, the first type packet is a BUM packet received by the receiving end node from an AC (access circuit, ATTACHMENT CIRCUIT) with Leaf attribute, and when the ESI of each ES ethernet automatic discovery route is not 0, the first type packet is a BUM packet received by the receiving end node from an AC with the same ESI.

In one exemplary embodiment, further comprising: and receiving a message carrying a copy of the second message and a label issued by the node through each ES (Ethernet) automatic discovery route, determining a first index according to the label carried in the message, acquiring a horizontal split label carried in the node through each ES (Ethernet) automatic discovery route according to the first index when forwarding the copy of the second message to other nodes, and sending the horizontal split label and the copy of the second message to the node together, wherein the first index is the label carried in the message or the first index is an index recorded in an input label mapping entry corresponding to the label carried in the message.

In one exemplary embodiment, further comprising: receiving a nineteenth route which has the same ESI value as the auto-discovery route of each ES, wherein the nineteenth route is the auto-discovery route of each ES and a route target carried in the nineteenth route is matched with an import route target of another local MAC-VRF, modifying a horizontal split label in the nineteenth route into a label which is locally distributed through the auto-discovery route of each ES, and then distributing the nineteenth route. Wherein, the nineteenth route is an auto-discovery route of each ES ethernet, and the nineteenth route carries the same ESI value as the auto-discovery route of each ES ethernet, and for clarity of description, the steps may be described as follows: a nineteenth route having the same ESI value as the first per ES ethernet auto-discovery route is received, the nineteenth route being a second per ES ethernet auto-discovery route and a route destination carried in the nineteenth route matches with an ingress route destination of another MAC-VRF local thereto, wherein the first and second are merely used to distinguish the per ES ethernet auto-discovery route as different routes, and are not particularly limited.

From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.

In another embodiment of the present invention, a cross-layer data packet sending device is provided, and the device is used to implement the foregoing embodiments and preferred embodiments, and will not be described in detail. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

Fig. 8 is a block diagram of a cross-layer data packet transmission apparatus according to an embodiment of the present invention, and as shown in fig. 8, a transmission apparatus 80 includes: a first receiving module 810, configured to receive a first packet from a second node, where the first packet includes a second packet;

And a first sending module 820, configured to determine whether a node corresponding to the third indication information meets a first condition, execute a first process if the node meets the first condition, and execute a second process if the node does not meet the first condition, where the first process is to prohibit sending the second message to the node corresponding to the third indication information, the second process is to send the second message to the node corresponding to the third indication information, where the first condition is that the first node receives a third route issued by the second node, and the third route is a route carrying the third indication information.

In another embodiment of the present invention, a routing packet transmission device is provided, fig. 9 is a block diagram of a routing packet transmission device according to an embodiment of the present invention, and as shown in fig. 9, a transmission device 90 includes: a second receiving module 910, configured to receive the first route; the second sending module 920 is configured to send a third route to the first node, where the third route carries third indication information, the third indication information is configured to instruct the first node to perform at least one of a first process and a third process in the third route, where the first process is to prohibit forwarding of the second message to a node that issues the first route corresponding to the third indication information, the third process is to perform DF election for the second node according to the third route, and determine whether to forward the fifth message to the node that issues the first route according to an election result of the DF election specific to the node that issues the first route, the fifth message is from a second layer node, the second layer node is a node other than the first cluster, and the first cluster is a cluster where the first node, the second node and the node that issues the first route are located.

In another embodiment of the present invention, there is provided a routing message retransmission apparatus, and fig. 10 is a block diagram of a routing message retransmission apparatus according to an embodiment of the present invention, and as shown in fig. 10, the retransmission apparatus 100 includes: and a resending module 1010, configured to receive the per-ES ethernet auto-discovery route, modify the horizontal split label in the per-ES ethernet auto-discovery route when an eighth condition is satisfied, and then issue the per-ES ethernet auto-discovery route, where the eighth condition is that a route target carried in the per-ES ethernet auto-discovery route matches with an import route target of the local MAC-VRF.

It should be noted that each of the above modules may be implemented by software or hardware, and for the latter, it may be implemented by, but not limited to: the modules are all located in the same processor; or the above modules may be located in different processors in any combination.

Embodiments of the present invention also provide a computer readable storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the method embodiments described above when run.

In one exemplary embodiment, the computer readable storage medium may include, but is not limited to: a usb disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a removable hard disk, a magnetic disk, or an optical disk, or other various media in which a computer program can be stored.

An embodiment of the invention also provides an electronic device comprising a memory having stored therein a computer program and a processor arranged to run the computer program to perform the steps of any of the method embodiments described above.

In an exemplary embodiment, the electronic apparatus may further include a transmission device connected to the processor, and an input/output device connected to the processor.

Specific examples in this embodiment may refer to the examples described in the foregoing embodiments and the exemplary implementation, and this embodiment is not described herein.

It will be appreciated by those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, they may be implemented in program code executable by computing devices, so that they may be stored in a storage device for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than that shown or described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.

In order to enable those skilled in the art to better understand the technical solutions of the present invention, the following description is provided with reference to specific exemplary embodiments.

In the following scenario embodiment, fig. 11 is a forwarding trace diagram of a second packet of the cross-layer data packet transmission method according to the scenario embodiment of the present invention, as shown in fig. 11, for the second packet of CE2 to be transmitted to VS3 node, VS3 node corresponds to the third node, when VS3 node forwards the second packet to VH2 but not to VH1, the second node may be VH2 node (on the other hand, if VS3 node transmits the second packet to VH1 but not to VH2, then VH1 node corresponds but not VH2 node any more), accordingly, VH1 node corresponds to the first node (on the other hand, if VS3 node transmits the second packet to VH1 but not to VH2, then VH2 node corresponds to VH1 node, and accordingly, node corresponding to the third layer node may be VS1 node, VS2 node, and VS3 node). Or if the CE2 sends the second message to the VS3 node but sends the second message to the VS2 node, the node corresponding to the third node is the VS2 node and no longer the VS3 node, and correspondingly, the nodes corresponding to the first node and the second node in the graph also change correspondingly.

Scene embodiment one

As shown in fig. 11, for the VH1 node, a first packet is received from VH2, where the first packet includes a second packet, the eleventh packet and the twelfth packet are respectively for different copies of the second packet, the eleventh packet is "broken for the ingress vSpoke node (corresponding loop is shown as" first loop "in fig. 11), the twelfth packet is" broken for the egress vSpoke node is indicated as ES2 (corresponding loop is shown as "second loop" in fig. 11) ", where the ingress vSpoke node is a node at which the second packet enters the first packet, the egress vSpoke node is a node at which the second packet can leave the first packet, the ES2 is an ES at which the second packet originates, the ingress vSpoke node is connected to the ES2, both VH1 and VH2 are not adjacent to the ES2," broken for the ingress vSpoke node "means that forwarding of the copy is prohibited when the designated copy of the second packet is the egress vSpoke node is the egress vSpoke node," forwarding the copy is prohibited "indicating that the egress vSpoke node is the ES2 is the ES at the point at which the point of the ingress 4225 is not adjacent to the designated ES2 from the designated ES2 node at the point at which the point of the ES2 is indicated as the point at which the point at the point of the ES 4225 is not being received from the ES2.

In the embodiment of the present invention, description is made for both data packet forwarding and routing packet forwarding, in the embodiment of the present invention, a disruption (i.e. loop disruption) flow in a data packet forwarding process is described first, and fig. 12 is a flow chart of a data packet forwarding principle according to the embodiment of the present invention, as shown in fig. 12, including:

Step S1202, VH1 receives a first message from VH2, where the first message includes a second message, the second message enters the first cluster from the VS3 node, the second message is derived from ES2, the first message further includes a first horizontal split tag, the first horizontal split tag is a horizontal split tag corresponding to ES2, and ES2 is an ethernet segment ES that is not adjacent to both VH1 and VH 2;

In step S1202, the format of the first packet may be an EVPN data packet, and the format of the second packet may be an ethernet packet.

In step S1204, VH1 determines a corresponding copy for each egress vSpoke node of the second packet, and performs eleventh and twelfth processing on the second packet. The copy of the second message corresponding to VS3 is denoted as p1_vs3, and the copy of the second message corresponding to VS2 is denoted as p1_vs2.

In step S1204, the eleventh process prohibits the forwarding of p1_vs3 (i.e., the copy of the second message generated after the second message enters the first cluster from the VS3 node) to the VS3 node (as shown by reference numeral ④ in fig. 11). The twelfth process is to indicate that the copy p1_vs2 of the second message by the VS2 node is an ES2 break (as shown by reference numeral ⑤ in fig. 11), where both VS2 and VS3 are connected to ES 2.

Note that, the eleventh process breaks the first loop by breaking the VS3 node (as shown in fig. 11), and the twelfth process breaks the second loop by indicating that the VS2 node breaks the ES2 loop (as shown in fig. 11), where the first loop is a loop inside the first cluster, and the second loop is a loop formed by the ethernet segment adjacent to the third layer node of the first cluster and the first cluster together.

For the data packet forwarding principle shown in fig. 11, in this embodiment of the present scenario, the following control plane, namely, a routing packet sending flow is proposed, and fig. 13 is a flowchart of a routing packet sending method according to an embodiment of the present invention, as shown in fig. 13, where the flow includes:

step S1302, VS3 issues a route r_vs3, VS2 issues a route r_vs2, wherein r_vs3 carries a thirteenth identifier, and r_vs2 carries a twelfth identifier;

Step S1304, in response to r_vs3 being validated, VH2 issues r_vh2_vs3 for VS3, wherein an identifier corresponding to VS3 is carried (in some cases, the value of which may be equal to the thirteenth identifier), and in response to r_vs2 being validated, VH2 issues r_vh2_vs2 for VS2, wherein an identifier corresponding to VS2 is carried (in some cases, the value of which may be equal to the twelfth identifier); in response to the failure of R_VS2, VH2 is VS2 to cancel R_VH2_VS2, wherein R_VH2_VS2 carries an identifier corresponding to VS2;

Wherein, in response to the failure of R_VS2, VH2 is a step S1404-5 for the specific process of removing R_VH2_VS2 for VS2.

Step S1306, VH1 receives r_vh2_vs3 from VH2, and notifies r_vh2_vs3 to the forwarding plane to form a third forwarding entry for VH1 to use when disruption for VS3 is made; wherein, still include: VH1 receives the withdrawal of r_vh2_vs2 from VH2, withdraws the second forwarding table entry corresponding to r_vh2_vs2, and the third forwarding table entry and the second forwarding table entry are different entries in the same forwarding table, so that the disruption to a designated node (such as VS 3) is to prohibit the message forwarded by the node from being sent to the node.

The third forwarding table entry corresponds to PEDLT (VH 2, VS 3) in fig. 17. The second forwarding table entry corresponds to PEDLT (VH 2, VS 2) in fig. 16. The third forwarding table entry and the second forwarding table entry are different entries in a same forwarding table, which may be, for example, a PE discriminator location table (PE Distinguisher Location Table, PEDLT). The message that has been forwarded through the node may be a message that enters a designated cluster from the node.

It should be noted that, the failure of r_vs2 means that r_vs2 is not used for forwarding a data packet, and the validation of r_vs2 means that r_vs2 is in a state that can be used for forwarding a data packet.

It should be noted that, for convenience of description, the present invention uses r_vhx_vsy to represent the route issued by the VHx node and containing the information corresponding to the VSy node; the R_VSy is used for representing the route issued by the VSy node; the unified R_VSy_SHL represents the route which is issued by the VSy node and carries a horizontal split label SHL (such as an ESI label or a Leaf label); the term R_VHx_ SHLy is used to denote the route of R_VSy_SHL after VHx reissue (R_VSy_SHL may be modified during reissue). Such as: r_vh2_vs3 represents a route issued by the VH2 node including information corresponding to the VS3 node, r_vs2 represents a route issued by the VS2 node, r_vs2_shl represents a route issued by the VS2 node carrying a horizontal split tag, and r_vh1_shl2 represents a route after re-issue of r_vs2_shl through VH 1.

Fig. 14 is a diagram of forwarding trajectories of a second packet of a cross-layer data packet sending method according to an embodiment of the present invention, as shown in fig. 14, further illustrating a data packet forwarding principle with respect to the data packet forwarding trajectories shown in fig. 14, where numbers ①～⑩ in fig. 12 correspond to steps S1401 to S1410, respectively.

In step S1401, the third node (VS 3) includes the second message from the first CE in the ethernet segment ES2 (i.e. the CE2 in the ethernet segment ES 2) and the second horizontal split label (i.e. the horizontal split label corresponding to the ES 2) in the fourth message, and sends the fourth message to the second node; the second node (VH 2) receives the second message and the second horizontal split tag from a third node (VS 3), and the second node determines a copy of the second message for each vSpoke node (including VS 1) other than VS3, wherein the copy determined for VS1 is denoted p2_vs1.

Step S1402, the second node (VH 2) encapsulates the second message in the first message, and sends the first message to the first node (VH 1), and the second node writes a tag (PEDL corresponding to PED of VH2 in this example) issued by the first node and identifying the second node in the first message; the first node (VH 1) receives a first message from a second node, wherein the first message comprises the second message, the second message enters the first cluster from a third node (VS 3), the first cluster is a cluster where the first node, the second node and the third node are located, the first node and the second node are first layer nodes (vHub nodes) in the first cluster, and the third node is a node (including vSpoke nodes (third layer nodes) in the first cluster) outside the first layer nodes.

In step S1403, when the second node (VH 2) determines that the copy of the second message determined for VS1 is p2_vs1, the third layer node corresponding to p2_vs1 (i.e., the egress node of p2_vs1) is VS1, and the second node receives the second message from the third node, the second node (VH 2) sends the second message (i.e., p2_vs1) to the fourth node (VS 1).

In step S1403, the fourth node belongs to the third layer node, but the fourth node and the third node are not the same node.

In step S1404, the first node (VH 1) marks a copy of the second message determined by VS2 (the node corresponding to id_vs2) as p1_vs2, and the first node determines whether the node corresponding to id_vs2 in the third layer node (i.e. the egress node of p1_vs2, that is, VS 2) satisfies the first condition corresponding to id_vs2, if so, the first process corresponding to id_vs2 is performed, and if not, the second process corresponding to id_vs2 is performed, wherein the third layer node is VS2, the first process corresponding to id_vs2 is to prohibit sending the second message (p1_vs2) to VS2, and the second process corresponding to id_vs2 is to decide whether to send the second message (p1_vs2) to VS2 according to the result of DF election specific to VS2, and if so, the first condition corresponding to id_vs2 is that the first node receives the third route corresponding to id_vs2 issued by the second node, and the third route corresponding to id_vs2 carries the information corresponding to VS 2.

In step S1405, the first node (VH 1) marks a copy of the second message determined by VS3 (the node corresponding to id_vs3) as p1_vs3, and the first node determines whether the node in the third layer node corresponding to id_vs3 (i.e. the egress node of p1_vs3), that is, VS 3) satisfies the first condition corresponding to id_vs3, if so, performs the first process corresponding to id_vs3, if not, performs the second process corresponding to id_vs3, wherein the first process corresponding to id_vs3 is to prohibit sending the second message (p1_vs3) to VS3, and the second process corresponding to id_vs3 is to decide whether to send the second message (p1_vs3) to VS3 according to the DF election result specific to VS3, wherein the first condition corresponding to id_vs3 is that the first node receives the third route corresponding to id_vs3 of the second node, and the third route corresponding to id_vs3 of the second node is indicated by the third route corresponding to VS 3.

In step S1406, the first node (VH 1) marks a copy of the second message determined by VS1 (the node corresponding to id_vs1) as p1_vs1, and the first node determines whether the node in the third layer node corresponding to id_vs1 (i.e. the egress node of p1_vs1, that is, VS 1) satisfies the first condition corresponding to id_vs1, if so, the first process corresponding to id_vs1 is executed, and if not, the second process corresponding to id_vs1 is executed, wherein the first process corresponding to id_vs1 is to prohibit sending the second message (p1_vs1) to VS1, and the second process corresponding to id_vs1 is to decide whether to send the second message (p1_vs1) to VS1 according to the DF election result specific to id_vs1, wherein the first condition corresponding to id_vs1 is that the first node receives the third route corresponding to id_vs1 of the second node, and the third route corresponding to id_vs1 is issued by the first node, and the third route corresponding to id_vs1 carries the information corresponding to VS 1.

In step S1407, the copy of the second message determined by the first node (VH 1) for CE4 is denoted p1_ce4, and VH1 forwards p1_ce4 to CE4.

It should be noted that, during the forwarding process of the second packet, the first node (VH 1) may respectively construct a copy of the second packet for each vSpoke node in the first cluster, where the copy constructed for VS1 is denoted as p1_vs1, the copy constructed for VS2 is denoted as p1_vs2, the copy constructed for VS3 is denoted as p1_vs3, and these copies are constructed before steps S1406, S1404, and S1405, respectively.

It should be noted that, during the forwarding process of the second packet, the second node (VH 2) constructs a copy of the second packet for each active vSpoke node except for VS3 in the first cluster, where the copy constructed for VS1 is denoted as p2_vs1 (after step S1404-5, VS2 is not considered as an active node and does not need to have a corresponding copy), and the p2_vs1 copy needs to be constructed before step S1403.

Fig. 18 is a schematic diagram of a location where a data packet is received/sent according to an embodiment of the present invention, where a first packet, a third packet, and a fourth packet carry copies of a second packet or a second packet, and a fifth packet and a sixth packet are different copies of a seventh packet respectively; the fourth message carries a second ESI label, the first message carries a first ESI label, the third message carries a third ESI label, and the eighth message carries a ninth message. Identical arrows are copies of the same message. The dashed line indicates that duplicate transmissions are prohibited.

Before step S1402, the method further includes the following steps:

in step S1402-1, the first node assigns PEDL a label to the second node.

In step S1402-2, the first node issues PEDL the label and its corresponding relationship with the second node to the second node.

In step S1402-3, the second node receives PEDL the tag.

Before step S1405, the method further includes the following steps:

Step S1405-1, VS3 issues route R_VS3, wherein thirteenth identifier is carried;

In step S1405-2, the second node receives R_VS3 from VS3.

In response to the validation of r_vs3, the second node sends a route r_vh2_vs3 (i.e., a third route corresponding to p1_vs3) to the first node in step S1405-3, wherein the route r_vh2_vs3 carries indication information id_vs3 (i.e., indication information corresponding to VS 3) for indicating the first node to perform at least one of a first process and a third process in the route r_vh2_vs3, wherein the first process is to prohibit forwarding of a second message Wen Fuben (i.e., p1_vs3) corresponding to VS3, the third process is to perform DF election according to the route r_vh2_vs3, and decide whether to forward a fifth message (as shown in fig. 18) to VS3 or not according to the election result of DF election specific to VS3, the second message is included in the first message sent to the first node in the route r_vh2_vs3, and the fifth message is from the second layer node, which is a node outside the first layer of the node.

In step S1405-4, the first node receives the route R_VH2_VS3.

Before step S1404, the method further includes the following steps:

step S1404-1, VS2 issues route R_VS2, wherein, the route R_VS2 carries twelfth identifier;

in step S1404-2, the second node receives R_VS2 from VS2.

In response to the validation of r_vs2, the second node sends a route r_vh2_vs2 (i.e., a third route corresponding to p1_vs2) to the first node, wherein the route r_vh2_vs2 carries indication information id_vs2 (i.e., indication information corresponding to VS 2), the indication information id_vs2 is used to indicate the first node to perform at least one of a first process and a third process in the route r_vh2_vs2, wherein the first process is to prohibit forwarding of a second message Wen Fuben (i.e., p1_vs2) corresponding to VS2, the third process is to perform DF election according to the route r_vh2_vs2, and decide whether to forward a fifth message to VS2 according to the election result of DF election specific to VS2, the second message is included in a first message sent from the second node to the first node, and the fifth message is from the second layer node, and the second layer node is a node outside the first cluster.

In step S1404-4, the first node receives a route R_VH2_VS2.

In step S1404-5, a fault occurs in the network causing r_vs2 to fail at the second node, which in response to r_vs2 failing, transmits the mp_ UNREACH _nlri of route r_vh2_vs2 to the first node.

In step S1404-6, the first node receives mp_ UNREACH _nlri, which is regarded as not receiving route r_vh2_vs2.

After steps S1404-6 and S1405-4, the process performed on p1_vs3 in step S1405 corresponds to the eleventh process, and the process performed on p1_vs2 in step S1404 corresponds to the twelfth process. That is, for a specified copy, only one of the first process or the second process can be performed thereon, but for different copies, the eleventh process and the twelfth process may be performed thereon, respectively. Performing the eleventh and twelfth processes on the second message means performing the eleventh and twelfth processes on different copies of the second message, respectively. Number ④ in fig. 11 is the result of the execution of number ④ in fig. 14 after step S1404-6. The dashed line in fig. 14 represents a copy of the second message that is prohibited from being forwarded after step S1404-6.

It should be noted that, in step S1404-5, the following step S1404-5-1 may be further included: in response to the R_VS2 failure, the reception of the second message from the VS2 node is stopped.

It should be noted that when r_vs1 or r_vs3 fails at the second node, the subsequent processing flow is similar to the corresponding processing flow in steps S1404-5 and S1404-6.

It should be noted that, because step S1404-6 precedes step S1404, in step S1404, the actual execution result of p1_vs2 by the first node will be to execute the second process, i.e. "decide whether to send the second message (p1_vs2) to VS2 according to the DF election result specific to VS2". If step S1404 occurs before step S1404-6 and after step S1404-4, accordingly, in step S1404, the actual execution of p1_vs2 by the first node results in the execution of the first process, i.e. "prohibit sending the second message (p1_vs2) to VS2". Thus, VH2 can control the forwarding behavior for p1_vs2 on VH1 by executing step S1404-5 as a trigger condition of step 1404-6 at an appropriate timing.

Before step S1406, the method further includes the following steps:

step S1406-1, VS1 issues a route R_VS1, wherein an eleventh identifier is carried;

in step S1406-2, the second node receives R_VS1 from VS1.

In response to the validation of r_vs1, the second node sends a route r_vh2_vs1 (i.e., a third route corresponding to p1_vs1) to the first node, wherein the route r_vh2_vs1 carries indication information id_vs1 (i.e., indication information corresponding to VS 1), the indication information id_vs1 is used to indicate the first node to perform at least one of a first process and a third process in the route r_vh2_vs1, wherein the first process is to prohibit forwarding of a second message Wen Fuben (i.e., p1_vs1) corresponding to VS1, the third process is to perform DF election according to the route r_vh2_vs1, and decide whether to forward a fifth message (as shown in fig. 18) to VS1 or not according to an election result of DF election, the second message included in a first message sent to the first node by the second node, the fifth message being from a node of the second layer, and the node of the second layer being a node outside the first cluster.

In step S1406-4, the first node receives a route R_VH2_VS1.

Step S1406-5, the first node performs DF election according to the route r_vh2_vs1, and the first node is a DF node specific to VS1 determined according to the DF election, performs DF election according to the route r_vh2_vs1, further comprising: the first node issues a route R_VH1_VS1 carrying indication information ID_VS1 corresponding to VS1 to the second node, and the first node performs DF election according to the routes R_VH2_VS1 and R_VH1_VS1.

After step S1406-5, there may be the following steps:

in step S1406-5-a1, the first node receives a fifth message from outside the first cluster (as shown in FIG. 18).

In step S1406-5-a2, the first node sends a fifth message to VS1 (as shown in fig. 18) according to the first node being a DF node specific to VS1 determined from DF elections.

After step S1406-5, there may be the following steps:

Step S1406-5-b0, the first node (VH 1) receives the route r_vs1 from VS1, responds to the validation of r_vs1, and sends the route r_vh1_vs1 to the first node, wherein the route r_vh1_vs1 carries indication information id_vs1 (i.e. indication information corresponding to VS 1), the indication information id_vs1 is used for indicating the second node to perform at least one of a first process and a third process in the route r_vh1_vs1, the first process is to prohibit forwarding of a copy of a ninth message corresponding to VS1 to the VS1, the third process is to perform DF election according to the route r_vh1_vs1 (see step S1406-5-b 2), and decides whether to forward a sixth message (see step S1406-5-b 4) to the VS1 according to the election result of DF election, the ninth message is included in an eighth message (see fig. 18) sent from the first node to the second node, and the third process is from the second node outside the second layer of the cluster.

In step S1406-5-b1, the second node receives route R_VH1_VS1.

Step S1406-5-b2, the second node performing DF election according to the route r_vh2_vs1, and the second node being a non-DF node specific to VS1 determined according to the DF election, performing DF election according to the route r_vh2_vs1, further comprising: the second node performs DF election according to routes r_vh2_vs1 and r_vh1_vs1.

In step S1406-5-b3, the second node receives a sixth message from outside the first cluster (as shown in fig. 18).

Step S1406-5-b4, the second node prohibiting sending the sixth message to VS1 according to the second node being a non-DF node specific to VS1 determined from DF elections by r_vh1_vs1 and r_vh1_vs2.

In step S1406-5-b5, the second node receives a tenth packet (as shown in fig. 18) from the CE3 from the VS2 node, where the tenth packet is a BUM packet, and the second node sends the tenth packet to the VS1. This step may precede step S1404-5.

Step S1406-5-b6, the first node receives a fifth message from outside the first cluster (as shown in fig. 18).

Step S1406-5-b7, according to the first node being a DF node specific to the VS1 determined according to DF elections performed by r_vh1_vs1 and r_vh1_vs2, the first node sends the fifth message to the VS1.

Step S1406-5-b8, the second node receives an eighth message (as shown in fig. 18) from the first node.

Step S1406-5-b9, the second node prohibits sending the ninth message (or a copy of the ninth message) to the VS1, where the ninth message is a message carried in the eighth message.

It should be noted that, when step S1403 precedes step S1406-5-b2, since DF elections have not been performed in common according to r_vh2_vs1 and r_vh1_vs1, the second node may be a DF node specific to id_vs1, and when step S1203 follows step S1406-5-b2, since DF elections have been performed according to r_vh2_vs1 and r_vh1_vs1, the second node may be a non-DF node specific to id_vs1, and forwarding of p2_vs1 in S1403 is not affected whether the second node is a DF node specific to id_vs1 or a non-DF node specific to id_vs1. Whether in step S1403 or steps S1406-5-b5, the second node forwards BUM messages (such as the second message or tenth message) from a particular vSpoke node to other vSpoke nodes (such as VS1, whether the second node is a DF node specific to id_vs1 or a non-DF node specific to id_vs 1).

It should be noted that, as shown in steps S1406-5-b4 and steps S1406-5-b7, only one of the fifth message and the sixth message is forwarded to VS1 at most, because at most only one of the first node and the second node is the DF node specific to VS1 determined according to DF elections performed by r_vh1_vs1 and r_vh2_vs1. Therefore, even if the fifth message and the sixth message are different copies of the seventh message (as shown in fig. 18), respectively, only one copy of the seventh message can reach the VS1 node at most, so that the problem that the duplicate copies of BUM messages from outside the first cluster (corresponding to the second layer node) are sent to the same vSpoke node (corresponding to one of the third layer nodes) by the vHub node (corresponding to the first layer node) is avoided.

After step S1402 and before steps S1404, S1405, and S1406, step S1402-7 is further included: the first node determines that the first message was received from the second node based on PEDL contained in the first message.

In step S1402, the following sub-steps may be further included:

In step S1402-9-1, the second node obtains the second horizontal split tag and the second message from the fourth message (see step S1401) and determines the value of the first horizontal split tag according to the second horizontal split tag, and the second node encapsulates the second message and the first horizontal split tag in the first message and sends the first message to the first node.

In step S1404, the following substeps S1404-9-1 may be further included: the first node adds a third horizontal split tag to the second message (p1_vs2) to obtain a third message, and the first node sends the third message to VS2, where the third horizontal split tag is used to instruct VS2 to prohibit forwarding of the second message to the first ethernet segment ES (i.e. ES 2), and the value of the third horizontal split tag is equal to the value of the first horizontal split tag or the value of the third horizontal split tag meets a third condition, where the third condition is that the value of the third horizontal split tag is determined according to the value of the first horizontal split tag (see step S1404-9-6 specifically).

After step S1404-9-1 and step S1407, the following steps may be further included:

in step S1408, VS2 receives the third message corresponding to p1_vs2 from the first node (VH 1), obtains the third horizontal split label (see step S1404-9-1 for the origin of the third horizontal split label) and the second message therefrom, and prohibits forwarding the second message to ES2 according to the third horizontal split label being the horizontal split label of ES2 (as shown by reference numeral ⑧ in fig. 9). By adding the third horizontal split tag to the third message in step S1404-9-1, the function of indicating VS2 to be ES2 for breaking in this step is performed.

Step S1409, VS2 receives the third message corresponding to p1_vs2 from the first node (VH 1), and obtains therefrom the third horizontal split label (see step S1404-9-1 for the origin of the third horizontal split label) and the second message, and VS2 allows forwarding the second message to CE3 when VS2 is the DF node of the ES where CE3 is located according to the fact that the third horizontal split label is not the horizontal split label of the ES where CE3 is located.

In step S1410, VS1 receives the second message from the second node (VH 2), because VS1 is the DF node of the ES where CE1 is located, VS1 forwards the second message to CE1.

It should be noted that, after step S1405-4 and before step S1405, the following step S1405-4-1 may be further included: the VH1 creates forwarding table entries PEDLT (VH 2, VS 3) from the r_vh2_vs3.

It should be noted that, by introducing step S1405-4-1, in step S1405, it is possible to efficiently determine that VS3 satisfies the first condition according to the presence of forwarding table item PEDLT (VH 2, VS 3), because forwarding table item PEDLT (VH 2, VS 3) has r_vh2_vs3 indicating that the first node has received the release of the second node (VH 2), and the r_vh2_vs3 carries indication information id_vs3 corresponding to the VS3.

It should be noted that, before step S1402-9-1, the following steps may be further included:

In step S1402-9-1-1, the second node receives the route r_vh1_shl2 (as shown by reference numeral ② in fig. 19) from the first node, r_vh1_shl2 automatically discovers the route for each ES ethernet, ESI of r_vh1_shl2 is ESI of the label ES2, wherein the first horizontal split label is carried, the second node modifies the next hop of r_vh1_shl2 and writes the second horizontal split label into r_vh1_shl2 as r_vh2_shl2, and re-distributes r_vh2_shl2 (as shown by reference numeral ③ in fig. 19). The r_vh2_shl2 is used to instruct the third node to add the second horizontal split flag in the fourth packet.

In step S1402-9-1-2, the third node receives r_vh2_shl2 (i.e., r_vh1_shl2 modified by the second node).

In step S1402-9-1-3, the third node uses r_vh2_shl2 to construct a fourth packet such that the fourth packet includes the second horizontal split flag.

It should be noted that, before step S1404-9-1, the following steps may be further included:

In step S1404-9-1-1, the first node receives a route r_vs2_shl from VS2 (as indicated by reference numeral ① in fig. 19), r_vs2_shl automatically discovers the route for each ES ethernet, ESI of r_vs2_shl is ESI of identity ES2, wherein a third horizontal split tag is carried, the first node modifies the next hop of r_vs2_shl and writes the first horizontal split tag into r_vs2_shl as r_vh1_shl2, and re-issues r_vh1_shl2 (as indicated by reference numeral ② in fig. 19). The r_vh1_shl2 is used to instruct the node receiving the r_vh1_shl2 to add a first horizontal split tag when sending a BUM data message to the first node according to the r_vh1_shl2.

In step S1404-9-1-2, the second node receives R_VH1_SHL2 (i.e., R_VS2_SHL modified by the first node).

In step S1404-9-1-3, the second node uses R_VH1_SHL2 to construct a first message such that the first message includes a first horizontal split tag.

It should be noted that, after step S1404-9-1-3, step S1404 may further include the following sub-steps:

in step S1404-9-6, when the first node receives the first message from the second node, the first node finds out the r_vh1_shl2 according to the horizontal split tag in the first message, and further finds out the third horizontal split tag carried before the r_vs2_shl is modified.

In step S1404-9-7, when the first node receives the first message from the second node, the first node sends the third horizontal split flag to VS2 along with the second message (i.e., p2_vs2).

It should be noted that, in this embodiment of the present scenario, the horizontal split tag may be an ESI tag.

Fig. 15 is a schematic flow chart of forwarding a routing message according to an embodiment of the present invention, as shown in fig. 15, wherein a number ① corresponds to steps S1406-1 and S1406-2; number ② corresponds to step S1406-3; number ③ corresponds to step S1406-4; number ④ has the same logic as number ④ in fig. 16, but corresponds to a different node; number ⑵ corresponds to step S1206-5-b 0. The second forwarding table entry corresponds to PEDLT (VH 2, VS 2) in fig. 15, PEDLT (VH 2, VS 2) and PEDLT (VH 2, VS 3) in fig. 17 are different forwarding table entries in the same forwarding table.

Fig. 16 is a schematic diagram of a routing message forwarding flow according to an embodiment of the present invention, as shown in fig. 16, in which a number ① corresponds to steps S1404-1 and S1404-2; number ② corresponds to step S1404-3; number ③ corresponds to step S1404-4; number ④ has the same logic as number ④ in fig. 16, but corresponds to a different node; number ⑵ has the same logic as number ⑵ in fig. 15, but corresponds to a different node.

Fig. 17 is a schematic diagram of a routing packet forwarding flow according to an embodiment of the present invention, as shown in fig. 17, where a number ① corresponds to steps S1405-1 and S1405-2; number ② corresponds to step S1405-3; the number ③ corresponds to step S1405-4; number ④ corresponds to step S1405-4-1; number ⑵ has the same logic as number ⑵ in fig. 15, but corresponds to a different node. As shown in fig. 17, the notification and revocation of forwarding table entries PEDLT (VH 2, VS 3) is described, where PED location table PEDLT is a forwarding table whose forwarding plane is used to query whether a certain PE (denoted PEx) issues an IMET route for another PE (denoted PEy) that carries a PED with an identifier PEy in PEDLA, and PEDLT (VH 2, VS 3) is a forwarding table entry in the PED location table that is used to query whether VH2 issues an IMET route for VS3 that carries a PED with an identifier VS3 in PEDLA. Issuing IMET (r_vh1_vs3) may refer to adding PED corresponding to VS3 to PEDLA of the IMET route issued by VH 1; dropping IMET (r_vh1_vs3) may refer to removing PED corresponding to VS3 in PEDLA of the IMET route issued by VH 1.

As shown in fig. 15, 16, and 17, in this scenario embodiment, each of r_vh1, VS1, r_vh2_vs1, r_vh1_vs2, r_vh2_vs2, r_vh1_vs3, and r_vh2_vs3 may be a inclusive multicast ethernet label IMET route. R_vs1, r_vs2 and r_vs3 can also be IMET routes.

It should be noted that, in the network of this embodiment of the present scenario, there may also be a BGP RR node (not shown in the figure), taking fig. 17 as an example, when VH1, VH2, VS1, VS2, and VS3 are routed through the same RR inter BGP route, the number ② and the number (2) refer to that the same route r_vh2_vs3 is sent to the RR node, because the RR node sends r_vh2_vs3 to VH1 and r_vh2_vs3 to VS3. The same applies to fig. 15 and 16.

It should be noted that PED tags can be issued by IMET routing carrying PEDLA as defined in draft-ietf-bess-evpn-virtual-hub-00. In steps S1402-1, S1402-2, two vHub of the same cluster are also assigned PEDL to each other and published in the IMET route PEDLA.

Scene embodiment two

The present scene embodiment differs from the first scene embodiment as follows:

In steps S1404-1, S1405-1 and S1406-1, R_VS1, R_VS2 and R_VS3 automatically discover routing for each EVI Ethernet, and in steps S1404-3, S1405-3 and S1406-3, ID_VS1, ID_VS2 and ID_VS3 are indication information corresponding to EVPN VPWS signaling instances EVPL_VS1, EVPL_VS2 and EVPL_VS3, respectively, which are EVPN VPWS signaling instances matching Ethernet tag identifiers ETI_VS1, ETI_VS2 and ETI_VS3 in R_VS1, R_VS2 and R_VS3, respectively.

In the case where the second node uses r_vs1 for forwarding a broadcast, unknown unicast or multicast BUM message, the second node prohibits using IMET route issued by VS1 for forwarding the BUM message. In the case where the second node uses r_vs2 for forwarding broadcast, unknown unicast or multicast BUM messages, the second node prohibits use of IMET routing issued by VS2 for forwarding BUM messages. In the case where the second node uses r_vs3 for forwarding broadcast, unknown unicast or multicast BUM messages, the second node prohibits use of IMET routing issued by VS3 for forwarding BUM messages.

The EVPN VPWS signaling instance evpl_vs1 is determined by l_eti_vs1 and eti_vs1, with id_vs1 being l_eti_vs1 or id_vs1 being id_evpl_vs1 corresponding to the EVPN VPWS signaling instance evpl_vs1. The EVPN VPWS signaling instance evpl_vs2 is determined by l_eti_vs2 and eti_vs2, with id_vs2 being l_eti_vs2 or id_vs2 being id_evpl_vs2 corresponding to the EVPN VPWS signaling instance evpl_vs2. The EVPN VPWS signaling instance evpl_vs3 is determined by l_eti_vs3 and eti_vs3, with id_vs3 being l_eti_vs3 or id_vs3 being id_evpl_vs3 corresponding to the EVPN VPWS signaling instance evpl_vs3.

It should be noted that l_eti_vs1, l_eti_vs2, l_eti_vs3 are local SERVICE INSTANCE IDENTIFIER (see RFC 8234) of evpl_vs1, evpl_vs2, and evpl_vs3 on the first node or the second node, respectively.

Bidirectional forwarding entities BFE corresponding to evpl_vs1, evpl_vs2 and evpl_vs3 are denoted as BFE1, BFE2 and BFE3, respectively, the second node binds BFE1, BFE2 and BFE3 to MAC-VRFs of the ethernet virtual private network EVPN, and BFE1, BFE2 and BFE3 are forwarding plane entities corresponding to respective EVPN VPWS signaling instances, respectively. VS3 indicates that the second message is from the first ES (i.e., ES 2) by adding a second horizontal split tag when sending the packet on BFE 3. In one embodiment, the horizontal split tag may be the flow tag of the BFE (i.e., the flow tag indicated in the EVPN VPWS signaling). In one embodiment, the horizontal split tag is obtained by looking up an ESI distribution table. The ESI distribution table is a forwarding table entry for querying forwarding information in the RT-1per ES route issued for the specified ESI by the specified PE.

It should be noted that in RFC8214, the BFE formed by the signaling instance of the EVPN VPWS is also for EVPN VPWS traffic at the forwarding plane, but BFE1, BFE2, and BFE3 herein are for only the corresponding EVPN VPWS signaling instance at the signaling plane, and are for MAC-VRF at the forwarding plane, not for EVPN VPWS traffic.

It should be noted that, the bidirectional forwarding entity in this embodiment of the present scenario is the same as the bidirectional forwarding entity of the EVPN VPWS service in RFC8214, except that in RFC8214, the bidirectional forwarding entity is used for the VPWS forwarding flow, and in this embodiment of the present scenario, is used for the MAC-VRF forwarding flow. The invention combines the forwarding plane entities in RFC8214 responsible for receiving packets from and sending packets to the tunnel as bidirectional forwarding entities, and transfers the bidirectional forwarding entities to the MAC-VRF. Therefore, the bidirectional forwarding entity in this embodiment of the present scenario may also be considered as a virtual interface.

It should be noted that the BFE masks ORIP (or next hop) the same IMET route in the ingress copy list as the far-end OPE TLV (or Peer IP) of the BFE.

For the present scene embodiment, id_evpl_vs1, id_evpl_vs2, and id_evpl_vs3 may be ESI and ETI, respectively, bound to the corresponding EVPN VPWS signaling instance. Accordingly, in this case, r_vh2_vs1, r_vh2_vs2, and r_vh2_vs3 may be auto discovery routes per EVI ethernet (also referred to as RT-1per EVI routes).

In step S1402, the second node encapsulates the first message according to the BFE corresponding to the tenth EVPN VPWS signaling instance, and steps S1402-1, S1402-2, S1402-3, and S1402-7 are not required.

Before step S1402, the method further includes the following steps:

Step S1402-21 configures the tenth EVPN VPWS signaling instance between the first node and the second node.

Step S1402-22 binds the tenth EVPN VPWS signaling instance to the EVPN MAC-VRF.

It should be noted that, after step S1402 and before steps S1404, S1405, S1406, steps S1402-27 are further included: and the first node determines that the first message is received from the second node according to tenth information contained in the first message, wherein the tenth information is an EVPN label carried in each EVI Ethernet automatic discovery route issued by the first node for the EVPN VPWS signaling example.

It should be noted that, in this embodiment of the present scenario, the route specifier RD and the route destination of the EVPN VPWS signaling instance may be configured separately, or may be shared with the EVPN signaling corresponding to the EVPN MAC-VRF.

It should be noted that, in this embodiment of the present scenario, the outbound label corresponding to the corresponding EVPN VPWS signaling instance needs to be written into the fourth packet, the first packet, and the third packet.

It should be noted that the present embodiment of the scenario implements vHub-vSpoke trunking functionality and enhances the failsafe capability of the vSpoke node without supporting IMET routing carrying PEDLA as defined in draft-ietf-bess-evpn-virtual-hub-00.

Scene embodiment III

In this embodiment of the present scenario, an example is given for a specific forwarding relationship and a packet label adding case in a forwarding process. And compared with the first embodiment of the scene, a different horizontal split label packaging mode and an outer layer tunnel packaging mode are provided.

For the encapsulation mode of the data message, an encapsulation mode different from the encapsulation mode can be adopted, for example, VXLAN encapsulation is adopted, a VXLAN tunnel from vHub nodes to vSpoke nodes can be a core copy tunnel, and a horizontal segmentation label-based tamper algorithm is adopted on VS1/VS2/VS3, wherein the horizontal segmentation label can be a global label (i.e. the horizontal segmentation label used by the same ESI on different adjacent nodes is the same label value).

In VXLAN EVPN, a horizontal split tag is carried in the UDP source port on the VXLAN outer layer. In one embodiment, the horizontal split tag is encrypted by the intrinsic entropy value, so that the original role of the UDP source port (e.g., for making load sharing routing uniform) can be still preserved to some extent while the horizontal split tag is transmitted. The receiving end can adopt the recovered intrinsic entropy value to decrypt the carried horizontal segmentation label value from the UDP source port.

Unlike "draft-ietf-bess-evpn-virtual-hub-00", when a vHub node sends a BUM message to the other vHub in the same cluster, the original source IP of the BUM message cannot be maintained unchanged, but the source IP is rewritten to the vHub node's own IP. Thus, the problem that the "draft-ietf-bess-evpn-virtual-hub-00" cannot support core replication due to the RPF check conflict with the underway multicast (such conflict caused by no overwriting of source IP) when the data plane is the VXLAN encapsulation is solved.

Scene example four

In this embodiment of the present scenario, an example is given for a specific forwarding relationship and a packet label adding case in a forwarding process. And a different third route distribution manner is given compared with the third embodiment of the scenario.

In this scenario embodiment, VH1, VH2 are AR-REPLICATOR nodes of the EVPN AR network, VS1, VS2, and VS3 are AR-LEAF nodes of the EVPN AR network, and the third route may be published through Replicator-AR routing messages defined in "draft-ietf-bess-EVPN-optimized-ir".

In step S1402, the second node encapsulates the first message according to the Reqular-IR route issued by the first node to the second node, and steps S1402-1, S1402-2, S1402-3, and S1402-7 are not required.

Before step S1402, the method further includes the following steps:

In step S1402-11, the first node issues Reqular-IR routes.

In step S1402-12, the second node receives the Regular-IR route.

After step S1402 and before steps S1404, S1405, and S1406, step S1402-8 is further included: the first node determines whether the first message is a message received from the first layer node according to information carried in the Regular-IR route contained in the first message, and only when the first message is a message received from the first layer node, the first condition is applicable to determine whether to execute the first process or the second process.

The active PE address list is carried in Replicator-AR routing (PEDLA may be employed). By adding or deleting IP addresses in the active PE address list, it is indicated whether other vHub nodes (referred to as AR-Replicator nodes in draft-ietf-bess-evpn-optimized-ir) are reachable from this vHub node to some vSpoke (referred to as AR-Leaf node in draft-ietf-bess-evpn-optimized-ir), where the presence of some PED indicates that this vHub is reachable to the corresponding vSpoke node, and the route receiver vHub node needs to prohibit forwarding of BUM messages received from this vHub node to this vSpoke node.

When PEDLA is employed to carry the active PE address list, in one embodiment only PED addresses are carried, and no valid PEDL is carried. In VXLAN EVPN, a horizontal split tag is carried in the UDP source port on the VXLAN outer layer. In one embodiment, the horizontal split tag is encrypted by the ground intrinsic entropy value, so that the original role of the UDP source port can be still preserved to some extent while the horizontal split tag is transmitted. The intrinsic entropy value, i.e. the entropy value that the receiving end can recover from other fields of the message (except for the field used to carry the intrinsic entropy value).

In this embodiment of the present scenario, the horizontal split tag may also be carried by Geneve/VXLAN-GPE encapsulation.

In this scenario embodiment, the BFE from the vHub node to the vSpoke node is a virtual channel with the VNI of the EVPN MAC-VRF as an in-label and the VNI in the IMET route of the vSpoke node as an out-label on the VXLAN tunnel from the vHub node to the vSpoke node. The source IP of the VXLAN tunnel is the AR-IP on the vHub node.

Scene embodiment five

In this embodiment of the present scenario, an example is given for a specific forwarding relationship and a packet label adding case in a forwarding process. And a different procedure is presented compared to the scene embodiment one.

Unlike the first embodiment of the present scenario, in this embodiment of the present scenario, there is also a vHub node VH3 (not shown in the figure), VH3 belonging to the same cluster as VH1 and VH 2.

In this embodiment, the VH2 sends the second message to VH3 when sending the first message to VH1, unlike the first embodiment.

Electing a DF specific to a given PED based on IMET routing with PEDLA, for a BUM message forwarded between vHub nodes, the BUM message can be sent to a node vSpoke only if the node vHub of the recipient of the BUM message is the DF specific to that node vSpoke. This is mainly the case when there are three or more vHub nodes in the same cluster (not shown in the figure). For example, after step S1404-6, in step S1404, when VH1 is a DF node specific to VS2, VH1 sends p1_vs2 to the VS2 node, and if VH3 is a DF node specific to VS2 and VH1 is a non-DF node specific to VS2, VH1 prohibits sending p1_vs2 to the VS2 node and VH3 sends p3_vs2 to the VS2 node, wherein p3_vs2 is the second message copy on VH3 corresponding to the VS2 node.

It should be noted that, rules for performing DF elections specific to a specific PED based on IMET routing can be summarized as follows: IMET routes published by the first layer node and carrying the specified PED (including corresponding IMET routes published by the current node) are ordered according to a specified rule, the node which publishes the first IMET route is determined to be DF node specific to the specified PED, and the node which publishes other IMET routes participating in the ordering is determined to be non-DF node specific to the specified PED. When multiple PEDs are carried in the same IMET route at the same time, each PED needs to be separately ordered and a DF election result specific to the corresponding PED is separately obtained.

It should be noted that, in the second process corresponding to p1_vs2 in step S1404 of the first scenario embodiment, the second packet (p1_vs2) may be sent to VS2 without judging the DF election result specific to VS2, because in the first scenario embodiment, there are only two nodes in the first layer node, and only one IMET route involved in the ranking in the second process remains, so that the DF node is the node performing the second process.

It should be noted that, by introducing DF elections specific to each vSpoke node respectively to the messages forwarded between vHub nodes, the problem that there may be multiple vHub nodes forwarding different copies of the second message to the same vSpoke node respectively is solved.

Scene example six

This embodiment is consistent with the scene embodiment except where specifically noted.

Unlike scenario embodiment one, this scenario embodiment merges r_vh2_vs1, r_vh2_vs2 and r_vh2_vs3 into the same route, which is denoted r_vh2_vs123 for convenience of description:

The second node sends to the first node the route r_vh2_vs123, which route r_vh2_vs123 may be an IMET route in which a list of < PEDs, PEDL > is carried by PEDLA, wherein id_vs1, id_vs2 and id_vs3 may be three different PEDs in the list, respectively, and the route r_vh2_vs123 is used to replace the role of r_vh2_vs1, r_vh2_vs2 and r_vh2_vs3 in scenario embodiment one.

Unlike the first embodiment of the scenario, the r_vh2_vs1 is sent by adding PED corresponding to VS1 to r_vh2_vs123; transmitting R_VH2_VS2 is to add PED corresponding to VS2 to R_VH2_VS123; the r_vh2_vs3 is transmitted by adding PED corresponding to VS3 to r_vh2_vs123.

Unlike scenario embodiment one, MP_ UNREACH _NLRI of BGP that publishes R_VH2_VS1 is replaced in this scenario embodiment with either the removal of PED corresponding to VS1 in PEDLA of R_VH2_VS123 or the removal of PEDL corresponding to PED of VS1 in PEDLA; mp_ UNREACH _nlri of BGP that issues r_vh2_vs2 is replaced in this scenario embodiment with PED corresponding to VS2 in PEDLA that removes r_vh2_vs123 or PEDL that removes PED corresponding to VS2 in PEDLA; mp_ UNREACH _nlri of BGP that issues r_vh2_vs3 is replaced in this scenario embodiment with the PED corresponding to VS3 removed in PEDLA of r_vh2_vs123 or PEDL corresponding to PED of VS3 removed in PEDLA.

When a specific PED is added to r_vh2_vs123, PEDL corresponding to the PED may be added.

Scene embodiment seven

This embodiment is consistent with scene embodiment six, except where specifically noted.

Unlike scenario six, the access circuits AC on ES2 in this scenario embodiment are all AC with E-Tree Leaf attributes.

Unlike scenario six, in this scenario embodiment CE1 is connected to VS1 through AC11 and to VS2 through AC21, both AC11 and AC21 having E-Tree Leaf attributes.

Unlike scenario embodiment six, in this scenario embodiment the ESI values of r_vh1_shl2, r_vs2_shl, and r_vs1_shl are 0, and the horizontal split group tags therein are carried by the E-Tree extended community attribute (such horizontal split group tags are also referred to as Leaf tags).

Unlike the sixth embodiment of the present scenario, in the embodiment of the present scenario, the second horizontal split tag in the fourth message in step S1401 is determined by the third node (VS 3) according to that the AC that receives the second message has the E-Tree Leaf attribute, and the AC is the AC of ES2 on VS3, so the second horizontal split tag is also the horizontal split tag determined according to ES2, and thus the second horizontal split tag is also the horizontal split tag corresponding to ES 2. Further, the first horizontal split tag in the first message and the third horizontal split tag in the third message are both determined directly or indirectly from the second horizontal split tag, and thus are also horizontal split tags corresponding to ES 2.

In this embodiment of the scenario, in step S1403-9-7, the fourth horizontal split tag sent by the second node to VS1 along with the second message (i.e., p2_vs1) is a tag in the E-Tree extended community attribute received by the second node from VS 1.

Before step S1403, the method may further include the following steps:

In step S1403-9-1-1, the second node receives the route r_vs1_shl from VS1 (as shown by reference numeral ① in fig. 20), r_vs1_shl is an auto-discovery route per ES ethernet with ESI value 0, wherein the fourth horizontal split label is carried, the second node modifies the next hop of r_vs1_shl and writes the fifth horizontal split label into r_vs1_shl as r_vh2_shl1, and then issues r_vh2_shl1 (as shown by reference numeral ② in fig. 20). The r_vh2_shl1 is used to instruct the node receiving the r_vh2_shl1 to add a fifth horizontal split tag when sending a BUM data message to the second node according to the r_vh2_shl1.

In step S1403-9-1-2, the third node receives R_VH2_SHL1 (i.e., R_VS1_SHL modified by the second node).

In step S1403-9-1-3, the third node uses R_VH2_SHL1 to construct a fourth message such that the fourth message includes a fifth horizontal split tag.

It should be noted that, after step S1403-9-1-3, step S1403 may further include the following sub-steps:

In step S1403-9-6, when the second node receives the fourth message from the third node, the second node finds the r_vh2_shl1 according to the horizontal split tag in the fourth message, and further finds the fourth horizontal split tag carried before the r_vs1_shl is modified.

In step S1403-9-7, when the second node receives the fourth message from the third node, the second node sends the fourth horizontal split tag to VS1 along with the second message (i.e., p2_vs1).

Note that, when the ESI values of r_vh1_shl1 (as shown by reference numeral ⑵ in fig. 20, where r_vs1_shl in reference numeral ⑴ is modified by the first node to be r_vh1_shl1) and r_vs1_shl are equal (for example, all are 0), the value of the twelfth horizontal split tag (i.e., the second node modifies r_vh1_shl1 to be horizontal split tag) and the value of the fifth horizontal split tag may also be equal. At this time, after step S1402-9-1-2 and step S1403-9-1-2, step S1403-9-6 may have the following sub-steps:

In step S1403-9-6-1, when the second node receives the fourth packet from the third node, the common ESI value of r_vh1_shl1 and r_vs1_shl is found according to the horizontal split tag (whether the twelfth horizontal split tag or the fifth horizontal split tag) in the fourth packet, and when forwarding the second packet Wen Fuben p2_vs1 corresponding to VS1, the second node finds r_vs1_shl for VS1 according to the ESI value and the egress node corresponding to p2_vs1, and further finds the fourth horizontal split tag carried before the r_vs1_shl is modified.

It should be noted that, according to the same logic, when forwarding the second message copy corresponding to the other node, a corresponding horizontal split label will be obtained.

Similarly, when the ESI values in r_vs2_shl (as shown by reference numeral ⑴ in fig. 19) and r_vh1_shl2 are also equal (e.g., both are labeled ES 2), a similar process can be performed.

It should be noted that, by executing the flow of updating the horizontal split label shown in step S1403-9-6-1 when forwarding the BUM data packet according to the received horizontal split label, the problem that the flow of updating the horizontal split label is implemented as label switching (Label Swapping) operation defined by RFC3031, which is caused by that only the corresponding egress node (for example, the egress node corresponding to the twelfth horizontal split label is the node issuing r_vh1_shl1, and the egress node corresponding to the fifth horizontal split label is the node issuing r_vs1_shl) can be avoided, and meanwhile, the problem of label waste caused by that different horizontal split labels are issued for r_vh1_shl1 and r_vs1_shl is solved.

Scene embodiment eight

This embodiment is seven in nature with the scene embodiment, except where specifically noted.

Unlike the seventh scenario embodiment, the eighth scenario embodiment also exists as follows: there are also cross-domain PE nodes in EVPN traffic where VS1, VS2, VS3, VH1, VH2 are located (for example, the second layer node may be a PE node of the same EVPN traffic in another AS domain), and ASBR in the Option B mode exists between the two AS domains.

Unlike scenario embodiment seven, because there is an eighth scenario, in this scenario embodiment the OPE TLVs defined in draft-heitz-bess-evpn-option-b are each carried in R_VS1_SHL, R_VS2_SHL, and R_VS3_SHL, which carry node identifiers for VS1, VS2, and VS3, respectively, which are consistent with ORIP in the corresponding IMET routes issued by VS1, VS2, and VS 3.

Unlike the seventh embodiment of the present scenario, in the present scenario embodiment, in step S1403-9-1-1, the following sub-steps are further included:

Step S1403-9-1-1, according to the matching of the route target carried in R_VS1_SHL and the EVPN import route target corresponding to the local MAC-VRF, writing the node identifier of the current node in the OPE TLV carried in R_VS1_SHL, wherein the written node identifier can cover the original initiator PE identification information in the OPE TLV.

It should be noted that the R_VH1_SHL in step S1402-9-1-1 and the R_VS2_SHL in step S1404-9-1-1 also have similar sub-steps as those of step S1403-9-1-1-1.

It should be noted that, when the virtual hub node and the virtual spoke node in the first cluster are implemented according to "draft-ietf-bess-evpn-virtual-hub" and "draft-heitz-bess-evpn-option-b", when there is an eighth situation, the transmission of the BUM message from CE3 to CE1 may be abnormal, specifically, both CE3 and CE1 are connected to the nodes in the first cluster through the AC with the E-Tree Leaf attribute, and according to the forwarding rule of the E-Tree service, the CE3 and CE1 should not receive the BUM message of each other, but the "draft-ietf-bess-evpn-virtual-hub" cannot guarantee this when there is the eighth situation. Because, in the eighth case, the OPE TLV cannot be modified at the time of route reissue according to the rule of "draft-heitz-bess-evpn-option-b", the horizontal split label in the fourth packet cannot be filled with the corresponding horizontal split label value, and thus the horizontal split labels in the subsequent first packet and third packet cannot be filled with the corresponding horizontal split label value, and thus the BUM packet forwarding from CE3 to CE1 is abnormal. By writing the node identifier of the current node in the OPE TLV carried in each ES Ethernet automatic discovery route before forwarding the each ES Ethernet automatic discovery route when the route target carried in the each ES Ethernet automatic discovery route is matched with the EVPN leading-in route target corresponding to the local MAC-VRF, the problem of abnormal BUM message forwarding from CE3 to CE1 in the first cluster when the method is implemented according to the 'draft-ietf-bess-EVPN-virtual-hub-00' and the 'draft-heitz-bess-EVPN-option-b' is solved.

Scene embodiment nine

This embodiment is consistent with scenario embodiment eight, except where specifically noted.

Unlike scenario eight, in this scenario embodiment, another method of modifying and reissuing ES-level ethernet auto-discovery routes is provided.

Unlike scenario embodiment eight, in the present scenario, when the ES-level ethernet auto-discovery route is modified by the next hop and reissued, the OPE TLV originally carried in the ES-level ethernet auto-discovery route is deleted.

Scene embodiment ten

This embodiment is consistent with scenario embodiment eight, except where specifically noted.

Unlike scenario eight, in this scenario embodiment, another method of modifying and reissuing ES-level ethernet auto-discovery routes is provided.

Unlike scenario embodiment eight, in the present scenario, when modifying the next hop for the ES-level ethernet auto-discovery route and reissuing the ES-level ethernet auto-discovery route, the OPE TLV originally carried in the ES-level ethernet auto-discovery route is not deleted or modified, but a new TLV is added, the TLV contains information identifying the current node, and the TLV instructs the receiving end node of the route to replace the role of the OPE TLV with the TLV.

Unlike scenario eight, in this scenario embodiment, the type of the new TLV corresponds to the second indication information and the value range of the new TLV corresponds to the seventh identifier.

Unlike the eighth embodiment of the present scene, in the present scene embodiment, further includes: and receiving a nineteenth route which has the same ESI value as the Ethernet automatic discovery route of each ES, wherein the nineteenth route is the Ethernet automatic discovery route of each ES, a route target carried in the nineteenth route is matched with an import route target of another local MAC-VRF, a horizontal split label in the nineteenth route is modified into a label issued by the node through the Ethernet automatic discovery route of each ES, and then the nineteenth route is issued.

Unlike the scene embodiment eight.

It should be noted that, by using the same horizontal split label value to modify the horizontal split label in the corresponding route when each ES ethernet with the same ESI value in different MAC-VRFs is reissued, the route receiving end node does not need to obtain the horizontal split label issued by the route transmitting end node according to different MAC-VRFs when transmitting the corresponding BUM message to the route transmitting end node, thereby saving the forwarding table, simplifying the forwarding flow, reducing the requirement on forwarding hardware, and solving the problem of wastage of the forwarding table.

In the above-described embodiment of the present invention, it is possible to release the restriction that the BUM message received from vHub must not be forwarded to the vSpoke node by breaking the loop. By introducing the first condition into the selection of the first process and the second process, the first loop is broken, and the use of a horizontal splitting label in the breaking flow of the first loop (shown in fig. 11) is avoided, so that the first layer node can transfer the horizontal splitting label for the ESI of the third layer node, the transferred horizontal splitting label can be used for indicating the third layer node to break the second loop (shown in fig. 11), the purpose of breaking the two layers of loops is achieved, and the CE of the third layer node can access the third layer node in a dual-homing/multi-homing mode. And then controlling whether the first condition can be met by controlling the release and withdrawal of the third route so that no loop is brought to the source vSpoke node while the VH1 node is forwarding the BUM message received from VH2 to other vSpoke nodes.

In summary, by introducing the selection procedure of the first process and the second process based on the first condition, the traffic between VH nodes can be filtered out from vSpoke nodes capable of safely receiving the traffic, and meanwhile, vSpoke nodes which may cause loops are excluded. And the screening or exclusion is not influenced by the Leaf tag carried in the message, the Leaf tag in the BUM message can be used in forwarding, but is not used in the screening (or exclusion), so that the Leaf tag can be used in an E-Tree forwarding flow on a third layer node, and the EVPN Hub-Spoke network can support EVPN E-Tree service. By introducing an OPE TLV rewriting mechanism, the EVPN Hub-Spoke network can be used in an E-Tree scene of an Option B cross-domain mode.

In summary, the present invention provides a method and apparatus for sending routing messages and cross-layer data messages, where when one vHub node is not reachable to a vSpoke node and another vHub node is reachable to the vSpoke node, the BUM packet received by the vSpoke node can still reach the vSpoke node by bypassing between two vHub nodes. The method can be applied to a scene that a plurality of vHub nodes exist in one Hub-Spoke cluster when Hub/Spoke deployment is carried out for EVPN, and is used for providing a protection path when one vHub node is not reachable to a certain vSpoke node, so as to avoid BUM messages to the vSpoke node from being discarded.

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention should be included in the protection scope of the present invention.

Claims (26)

1. The cross-layer data message sending method is characterized by comprising the following steps:

the method comprises the steps that a first node receives a first message from a second node, wherein the first message comprises a second message;

The first node judges whether a node corresponding to third indication information meets a first condition, if yes, first processing is executed, if not, second processing is executed, wherein the first processing is to prohibit sending of the second message to the node corresponding to the third indication information, the second processing is to send the second message to the node corresponding to the third indication information, the first condition is that the first node receives a third route issued by the second node, and the third route is a route carrying the third indication information.

2. The method of claim 1, wherein the third indication information is indication information corresponding to a node in a third layer of nodes, the third layer of nodes is composed of nodes other than a first layer of nodes in a first cluster, the first cluster is a cluster in which the first node and the second node are located, and the first layer of nodes is a node in which the first node and the second node are located.

3. The method of claim 1, further comprising, prior to said performing the second process:

and deciding whether to execute the second process according to DF election results specific to the node corresponding to the third indication information.

4. The method of claim 1, wherein prior to the first node receiving the first message from the second node, the method further comprises:

The first node issues a first label to the second node, the first label is included in the first message, and the first label is a label according to which the first node determines that the first message comes from the second node.

5. The method of claim 1, wherein the sending the second message to the node corresponding to the third indication information comprises:

The first message further comprises a virtual extended local area network VXLAN encapsulation, the first node obtains the first horizontal segmentation tag from a UDP source port of the VXLAN encapsulation of the first message, and the first node performs VXLAN encapsulation on the second message to obtain the third message, wherein a source IP of a tunnel of the VXLAN is the first node, and the UDP source port of the VXLAN carries the third horizontal segmentation tag.

6. The routing message sending method is characterized by comprising the following steps:

The second node receives the first route;

The second node sends a third route to the first node, wherein the third route carries third indication information, the third indication information is used for indicating the first node to execute at least one of first processing and third processing in the third route, the first processing is a node which prohibits forwarding of a second message to the first route, the node corresponds to the third indication information, the third processing is a node which performs DF election according to the third route, and decides whether to forward a fifth message to the node which issues the first route according to an election result of the DF election, the fifth message is from a second layer node, the second layer node is a node outside a first cluster, and the first cluster is a cluster where the first node, the second node and the node which issues the first route are located.

7. The method as recited in claim 6, further comprising:

The node issuing the first route is a node in a third layer of nodes in the first cluster, the third layer of nodes consists of nodes except the first layer of nodes in the first cluster, and the first layer of nodes are nodes of layers where the first node and the second node are located.

The second node receives the second message, and the second node sends the second message to a fourth node in the third layer node under the condition that the second message received by the second node comes from the third node in the third layer node.

8. The method of claim 6 or 7, wherein,

The second node performs DF election according to at least the third route, and the second node is a non-DF node determined according to the DF election that is specific to the node that issued the first route.

9. The method as recited in claim 8, further comprising:

And the second node receives a sixth message from the second layer node, and the second node prohibits the sixth message from being sent to the node issuing the first route.

10. The method of claim 8, wherein the second node performs DF elections according to at least the third route, comprising:

And the second node receives a sixth route which is issued by the first node and carries the third indication information, and the first node performs DF election according to the third route and the sixth route.

11. The method of claim 10, wherein the step of determining the position of the probe comprises,

The third route and the sixth route include an operator edge device identifier tag attribute PEDLA, the PEDLA carries a PE identifier PED of an operator edge device PE, the PED identifier is used as the third indication information, the PED identifier is used to identify the node that issues the first route, and the DF election determined DF node is specific to the PED identifier.

12. The method of claim 6, wherein the first route comprises one of:

an inclusive multicast ethernet label IMET route, wherein an internet address ORIP of an initiator router in the IMET route carries the third indication information;

and each EVI Ethernet automatic discovery route, wherein the third indication information is indication information corresponding to a first EVPN VPWS signaling example, and the first EVPN VPWS signaling example is an EVPN VPWS signaling example matched with a first Ethernet label identifier ETI in the each EVI Ethernet automatic discovery route.

13. The method of claim 6, wherein the first route includes the third indication information identifying the node that issued the first route in the first route.

14. The method of claim 12, wherein the first EVPN VPWS signaling instance is determined by the first ETI and a second ETI, the third indication information is the second ETI or the third indication information is fourth indication information corresponding to the first EVPN VPWS signaling instance.

15. The method of claim 6, wherein the second node sends a first message to the first node, and wherein the second node writes a tag issued by the first node that identifies the second node in the first message.

16. The method of claim 6, wherein the second node sends a third route to the first node, comprising:

the third route includes a PED label attribute PEDLA, the PEDLA carries a PE identifier PED corresponding to the node that issues the first route, and the second node sets a PED label PEDL in the PEDLA corresponding to the PED to a first specified value.

17. The method of claim 6, wherein, where the first route is an auto-discovery route per EVI ethernet, and where the first route is used to forward broadcast, unknown unicast, or multicast BUM messages, the second node prohibits using the IMET route issued by the node issuing the first route for forwarding BUM messages.

18. A method for resending a routing message, comprising:

And receiving the automatic discovery route of each ES, and when an eighth condition is met, modifying the horizontal division label in the automatic discovery route of each ES, and then releasing the automatic discovery route of each ES, wherein the eighth condition is that a route target carried in the automatic discovery route of each ES is matched with a leading-in route target of a local MAC-VRF.

19. The method as recited in claim 18, further comprising:

when the eighth condition is satisfied, performing one of the following processing on the per ES ethernet automatic discovery route:

Deleting an initiator provider edge OPE type length value TLV carried in the per-ES Ethernet automatic discovery route;

writing first indication information into the per-ES Ethernet automatic discovery route, wherein the first indication information indicates that an OPE TLV carried in the per-ES Ethernet automatic discovery route is ignored;

covering the initiator PE identification information in the OPE TLV carried in the automatic discovery route of each ES by using local node identification information;

Writing second indication information and a seventh identifier into the per-ES ethernet automatic discovery route, where the second indication information indicates that a receiving end node of the per-ES ethernet automatic discovery route adds a horizontal split label in the per-ES ethernet automatic discovery route when forwarding a first type packet according to a first type route, where the first type route is ORIP a provider multicast service interface I-PMSI route or a provider multicast service interface S-PMSI route of the seventh identifier, and when the ESI of the per-ES ethernet automatic discovery route is 0, the first type packet is a BUM packet received by the receiving end node from an AC with a Leaf attribute, and when the ESI of the per-ES ethernet automatic discovery route is not 0, the first type packet is a BUM packet received by the receiving end node from an AC with the same ESI.

20. The method as recited in claim 19, further comprising:

And receiving a message carrying a copy of a second message and a label issued by the node through the automatic discovery route of each ES (Ethernet), determining a first index according to the label carried in the message, acquiring a horizontal split label carried in the automatic discovery route of each ES by the node according to the first index when forwarding the copy of the second message to other nodes, and sending the horizontal split label and the copy of the second message to the node together, wherein the first index is the label carried in the message or the first index is an index recorded in an input label mapping entry corresponding to the label carried in the message.

21. The method as recited in claim 19, further comprising:

Receiving a nineteenth route which has the same ESI value as the auto-discovery route of each ES, wherein the nineteenth route is the auto-discovery route of each ES and a route target carried in the nineteenth route is matched with a local imported route target of another MAC-VRF, modifying a horizontal split label in the nineteenth route into a label which is locally distributed through the auto-discovery route of each ES, and then distributing the nineteenth route.

22. A cross-layer data message transmitting apparatus, comprising:

The first receiving module is used for receiving a first message from a second node, wherein the first message comprises a second message;

The first sending module is configured to determine whether a node corresponding to the third indication information meets a first condition, execute a first process if the node meets the first condition, and execute a second process if the node does not meet the first condition, where the first process is to prohibit sending the second message to the node corresponding to the third indication information, and the second process is to send the second message to the node corresponding to the third indication information, where the first condition is that the first node receives a third route issued by the second node, and the third route is a route carrying the third indication information.

23. A routing message transmitting apparatus, comprising:

a second receiving module for receiving the first route;

The second sending module is configured to send a third route to the first node, where the third route carries third indication information, where the third indication information is used to instruct the first node to perform at least one of a first process and a third process in the third route, where the first process is a node that prohibits forwarding of a second packet to the first route, which corresponds to the third indication information, and the third process is that the second node performs DF election according to the third route, and decides whether to forward a fifth packet to the first route, according to an election result of the DF election specific to the first route, where the fifth packet is from a second layer node, where the second layer node is a node outside a first cluster, and where the first node, the second node, and the first route are located.

24. A routing message resending apparatus, comprising:

And the retransmission module is used for receiving the automatic discovery route of each ES, modifying the horizontal segmentation label in the automatic discovery route of each ES when an eighth condition is met, and then releasing the automatic discovery route of each ES, wherein the eighth condition is that a route target carried in the automatic discovery route of each ES is matched with a local MAC-VRF import route target.

25. A computer readable storage medium, characterized in that a computer program is stored in the computer readable storage medium, wherein the computer program, when executed by a processor, implements the method of any of claims 1 to 21.

26. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any one of claims 1 to 21 when executing the computer program.