Communication method and apparatus

Abstract

This application provides a communication method and apparatus. The method is applied to a first PE, which is located within an EVPN network. The EVPN network also includes a second PE. The CE establishes connections with the first PE and the second PE through different links, respectively. The first PE and the second PE form a multi-homed group. The method includes: when the second PE fails, obtaining an ESI from the EAD per ES route sent by the second PE, wherein the ESI is an identifier of the same ES associated with the first PE and the second PE; obtaining a MAC-IP route sent by the second PE including the ESI based on the ESI; obtaining a first MAC entry generated corresponding to the MAC-IP route based on the MAC-IP route; and converting the first MAC entry into a second MAC entry.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a communication method and apparatus. Background Technology

[0002] In existing Virtual Private LAN Service (VPLS) networks, RFC 7432 proposes a scheme for fast unicast traffic convergence, which achieves fast unicast traffic convergence by canceling EAD per ES routes.

[0003] like Figure 1 As shown, Figure 1 This is a schematic diagram of an existing VPLS network architecture. Figure 1 In this architecture, a point-to-multipoint Ethernet Virtual Private Network (EVPN) is formed between Provider Edge (PE) 1, PE2, and PE3 to provide Layer 2 communication between Customer Edge (CE) 1 and CE2. The forwarding plane between PE1, PE2, and PE3 can employ Virtual Extensible Local Area Network (VXLAN) encapsulation, Multi-Protocol Label Switching (MPLS) encapsulation, or Segment Routing IPv6 (SRv6) encapsulation. CE1 accesses the EVPN via a dual-homed Ethernet segment (ES).

[0004] Once a PE learns the local Media Access Control Address (MAC) (e.g., the MAC of a virtual machine connected to the PE, the MAC of an Ethernet switch, or the MAC of a virtual machine connected to the PE via an Ethernet switch), it will synchronize the user-side MAC to other PEs with which it has established Border Gateway Protocol (BGP) neighboring relationships via the EVPN's MAC-IP routing (also known as MAC-IP Advertisement Route). For example, PE2 learns CE1's MAC1 from the AC connected to CE1 and synchronizes it to PE1 and PE3. After receiving the MAC-IP route advertised by PE2, PE3 generates a MAC entry for MAC1. If CE1's ES mode in the EVPN is single-active, the outgoing direction of the MAC entry points to the primary / backup path between PE1 and PE2; if CE1's ES mode in the EVPN is multi-active, the outgoing direction of the MAC entry points to the equivalent path between PE1 and PE2. Similarly, on PE1, the outgoing direction of the MAC entry points to the AC connected to CE1.

[0005] When the link between PE2 and CE1 fails, on PE3, if the ES mode is multi-active ES, PE3 needs to update the outgoing direction of the MAC table entry from the equivalent path pointing to PE1 and PE2 to the single path pointing to PE2; if the ES mode is single-active ES, PE3 needs to update the outgoing direction of the MAC table entry from the primary backup path pointing to PE1 and PE2 to the backup path.

[0006] Because the number of user-side MAC addresses learned by the PE is enormous, waiting for other PEs to complete the synchronization of all MAC-IP route updates will result in a large number of traffic packets being lost due to outgoing direction errors. Therefore, RFC7432 proposes a scheme for fast convergence of unicast traffic.

[0007] In other words, when a PE is associated with an ES, it needs to advertise an EVPN EAD per ES route. This route includes the Ethernet Segment Identifier (ESI) and the address of the PE advertising the route. Simultaneously, the MAC-IP route advertised by the PE also includes the ESI and the user-side MAC address of the ES indicated by the access ESI. For example, after PE3 receives the MAC-IP route advertised by PE2 and the EAD per ES route advertised by PE1 and PE2, the outbound direction of the MAC entry varies depending on the ES mode. When the link between PE2 and CE1 fails, the EAD per ES route advertised by PE2 is revoked, and all traffic including MAC1 can no longer be forwarded to PE2. Thus, only one route revocation is needed to modify the outbound direction of the MAC entry.

[0008] However, the above-mentioned fast unicast traffic convergence scheme exposes a problem when PE2 restarts. When PE2 restarts, all BGP neighbors between PE2 and PE1 and PE3 are lost. On PE1 and PE3, this means that all routes advertised by PE2 are unreachable (including EAD per ES routes and MAC-IP routes). Therefore, on PE3, not only is the outbound direction of the MAC table updated after the EAD per ES route is withdrawn, but the corresponding MAC table entry is also deleted after the MAC-IP route is withdrawn.

[0009] Similarly, the MAC entry will be deleted on PE1, causing unicast traffic on PE1 and PE3 to change to unknown unicast traffic. This continues until CE1 detects the PE2 failure and switches the traffic to PE1. PE1 then relearns the user-side MAC and publishes the traffic to PE3, at which point the traffic changes back to known unicast traffic.

[0010] In practical applications, to prevent excessive broadcast, unknown unicast, and multicast (BUM) traffic from overloading network-side devices, BUM traffic is typically rate-limited. Therefore, when a PE (Preinstallation Equipment) fails, significant unicast packet loss occurs. Consequently, a technique is urgently needed to reduce unicast packet loss in the event of a single node failure within a multi-homed group.

[0011] The existing draft, draft-eastlake-bess-enhance-evpn-all-active-08, proposes a delayed deletion scheme that can solve the aforementioned problems. Combined with... Figure 1 The main process is as follows:

[0012] PE1 and PE2 are both associated with the same Elasticsearch Entity (ES) and advertise EAD per EVI and EAD per ES routes to other PEs that have established BGP neighbor relationships. PE2 learns CE1's MAC address (MAC1) and synchronizes it with PE1 and PE3 via MAC-IP routes. After receiving the EAD per EVI and EAD per ES routes advertised by PE1 and PE2, as well as the MAC-IP route advertised by PE2, PE3 determines that CE1's ES access mode to the EVPN is multi-active ES. Therefore, the outgoing direction of the MAC entry in the locally generated MAC1 table is the equivalent path pointing to PE1 and PE2.

[0013] After receiving the EAD per EVI route, EAD per ES route, and MAC-IP route advertised by PE2, PE1 generates a MAC entry for MAC1 locally. The outgoing direction of the MAC entry points to the AC connected to CE1. At the same time, PE1 also advertises the MAC-IP route for MAC1.

[0014] After PE3 receives the MAC-IP route advertised by PE1, it determines that it is the same as the MAC-IP route advertised by PE2. PE3 will preferentially use one of the MAC-IP routes. Similarly, PE2 will also receive the MAC-IP route advertised by PE1. However, for PE2, since the MAC-IP route advertised by PE1 is the same as its own advertised MAC-IP route, PE2 will preferentially use the MAC1 route it learned, and will no longer process the MAC-IP route advertised by PE1.

[0015] When PE2 fails, all BGP neighbors between PE2 and PE1 and PE3 are disconnected. PE1 and PE3 will revoke all routes advertised by PE2. When PE1 processes the revocation of MAC-IP routes advertised by PE2, if it determines that the outgoing direction of the MAC entry corresponding to the MAC-IP route points to the access circuit (AC) connected to CE1, PE1 will not delete the MAC entry, nor will it revoke its own advertised MAC-IP routes. Instead, it will start a timer (e.g., 2 minutes). When PE3 processes the revocation of MAC-IP routes advertised by PE2, if it determines that it has also received the same MAC-IP route from PE1, PE3 will choose to keep the MAC-IP routes advertised by PE1 valid and will not delete the corresponding MAC entries.

[0016] After CE1 detects a PE2 failure, it switches traffic to PE1. Once PE1 relearns MAC1, it stops the timer. After the timer expires, PE1 deletes the MAC entries it has not learned and cancels the corresponding MAC-IP routes.

[0017] However, the delayed deletion scheme provided by draft-eastlake-bess-enhance-evpn-all-active-08 also exposes the following problems: 1) When the number of user-side MAC addresses is large, PE1 cannot guarantee that it will learn all user-side MAC addresses within the time limit. If not all user-side MAC addresses have been learned after the time limit, PE1 will still delete the MAC entries of the unlearned user-side MAC addresses, causing unicast traffic to change to unknown unicast traffic on PE1 and PE3; 2) It cannot distinguish between fault scenarios and aging scenarios. When the user-side MAC addresses learned by PE2 age normally, PE2 revokes the published MAC-IP routes. At this time, PE1 will still enter the delayed deletion scheme process, causing the aged user-side MAC addresses to continue to remain in PE1 and PE3. Summary of the Invention

[0018] In view of this, this application provides a communication method and apparatus to solve the problem in the existing draft delayed deletion scheme that when the number of user-side MAC addresses is huge, the PE cannot guarantee that it will learn all user-side MAC addresses within the time interval and that there are residual user-side MAC addresses in the PE that need to be aged.

[0019] In a first aspect, this application provides a communication method applied to a first PE, which is located within an EVPN network. The EVPN network further includes a second PE. The CE establishes connections with the first PE and the second PE respectively through different links, and the first PE and the second PE form a multi-homed group. The method includes:

[0020] When the second PE fails, the ESI is obtained from the EAD per ES route sent by the second PE. The ESI is the identifier of the same ES associated with the first PE and the second PE.

[0021] Based on the ESI, obtain the MAC-IP route including the ESI sent by the second PE;

[0022] Based on the MAC-IP route, obtain the first MAC entry generated corresponding to the MAC-IP route;

[0023] Convert the first MAC entry to the second MAC entry.

[0024] Secondly, this application provides a communication device applied to a first PE, the first PE being located within an EVPN network, the EVPN network further including a second PE, the CE establishing connections with the first PE and the second PE respectively through different links, the first PE and the second PE forming a multi-homed group, the device comprising:

[0025] The first acquisition unit is used to acquire the ESI from the EAD per ES route sent by the second PE when the second PE fails, wherein the ESI is the identifier of the same ES associated with the first PE and the second PE;

[0026] The second acquisition unit is used to acquire the MAC-IP route including the ESI sent by the second PE according to the ESI;

[0027] The third acquisition unit is used to acquire the first MAC entry generated corresponding to the MAC-IP route based on the MAC-IP route;

[0028] A conversion unit is used to convert the first MAC entry into a second MAC entry.

[0029] Thirdly, this application provides a network device including a processor and a machine-readable storage medium storing machine-executable instructions that can be executed by the processor, which in turn cause the processor to perform the method provided in the first aspect of this application.

[0030] Therefore, by applying the communication method and apparatus provided in this application, when the second PE fails, the ESI is obtained from the EADper ES route sent by the second PE. The ESI is the identifier of the same ES associated with the first PE and the second PE. According to the ESI, the first PE obtains the MAC-IP route sent by the second PE including the ESI. According to the MAC-IP route, the first PE obtains the first MAC entry generated corresponding to the MAC-IP route. The first PE converts the first MAC entry into the second MAC entry.

[0031] Therefore, there is no need to start an additional timer to delete synchronously learned MAC entries. When MAC entries age normally, since the EAD per ES route has not been revoked, PE2 and PE3 receive the MAC-IP route revocation and can delete the MAC entries normally, avoiding the long-term occupation of MAC table resources by useless MAC entries. When handling EAD per ES route revocation, the MAC-IP routes synchronized by PEs associated with the same ES are relearned as local MAC entries. Even if the number of user-side MACs is huge and the PE's learning speed of user-side MACs is slow, there will be no packet loss due to MAC entry deletion. This solves the problem in the existing draft's delayed deletion scheme where, when the number of user-side MACs is huge, PE cannot guarantee that it will learn all user-side MACs within the time limit and there are residual user-side MACs in PE that are waiting to age. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of an existing VPLS network structure;

[0033] Figure 2 A flowchart illustrating the communication method provided in the embodiments of this application;

[0034] Figure 3 A structural diagram of a communication device provided in an embodiment of this application;

[0035] Figure 4 The network device hardware structure provided in the embodiments of this application. Detailed Implementation

[0036] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0037] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the corresponding listed items.

[0038] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0039] The following is a detailed description of a communication method provided by an embodiment of this application. See also... Figure 2 , Figure 2 A flowchart illustrating a communication method provided in an embodiment of this application. This method is applied to a first PE. The communication method provided in an embodiment of this application may include the following steps.

[0040] Step 210: When the second PE fails, obtain the ESI from the EAD per ES route sent by the second PE. The ESI is the identifier of the same ES associated with the first PE and the second PE.

[0041] Specifically, the first PE is within the EVPN network, which also includes a second PE and a third PE. The third PE can be a remote PE. The CE establishes a connection with the first PE through the first link, and the CE establishes a connection with the second PE through the second link. The first PE and the second PE form a multi-homed group. The set of the first link and the second link is called the ES, and this set is identified by the ESI.

[0042] After successful network setup, both the first and second PEs are associated with the same Elasticsearch (ES) and advertise EAD per EVI and EAD per ES routes to other PEs that have established BGP neighbor relationships. When traffic reaches the second PE, the second PE learns the user-side MAC address and synchronizes it with the first and third PEs via the EVPN's MAC-IP routing.

[0043] It should be noted that EAD per EVI routes and EAD per ES routes can be collectively referred to as Ethernet Auto-discovery Routes (EAD routes), which are a type of route in EVPN. This type of route includes two sub-routes: EAD per EVI (EVPN Instance) routes and EAD per ES routes. EAD per EVI routes are used to discover primary, backup, or equal-cost paths within an EVPN. EAD per ES routes are used for rapid MAC table convergence in the event of an EVPN failure.

[0044] MAC-IP routing, also known as MAC-IP advertising routing, is a type II routing in EVPN, used to advertise MAC addresses and host routing information (i.e., ARP and ND information).

[0045] After receiving the EAD per EVI route and EAD per ES route advertised by the first PE and the second PE, as well as the MAC-IP route advertised by the second PE, the third PE determines that the ES mode of the CE accessing the EVPN is multi-active ES, and then generates a MAC entry for the user-side MAC on its local machine. The outgoing direction of this MAC entry is the equivalent path pointing to the first PE and the second PE.

[0046] After receiving the EAD per EVI route, EAD per ES route, and MAC-IP route advertised by the second PE, the first PE generates a MAC entry for the user-side MAC address locally. The outgoing direction of this MAC entry points to the AC connected to the CE. Simultaneously, the first PE also advertises the MAC-IP route for the user-side MAC address. This MAC-IP route is identical to the one advertised by the second PE.

[0047] After receiving the MAC-IP route advertised by the first PE, the third PE determines that it is the same as the MAC-IP route advertised by the second PE, and the third PE preferentially selects one of the MAC-IP routes to take effect. However, regardless of whether the route advertised by the first PE or the second PE is selected, the outgoing direction of the MAC entry generated by the third PE is the same.

[0048] The second PE will also receive MAC-IP routes advertised by the first PE. However, since the MAC-IP routes advertised by the first PE are the same as those advertised by the second PE, the second PE will prefer to use the user-side MAC addresses it has learned, and will no longer process the MAC-IP routes advertised by the first PE.

[0049] When the second PE fails, the first PE detects that its BGP neighbor relationship with the second PE has been broken. The first PE then triggers a process to revoke all routes synchronized with the second PE.

[0050] In the routing synchronized by the second PE, the EAD per ES route needs to be processed first because it has a fast convergence function. The first PE obtains the EAD per ES route sent earlier by the second PE from its local machine and obtains the ESI from the EAD per ES route. This ESI is the identifier of the same ES associated with the first PE and the second PE.

[0051] Step 220: Based on the ESI, obtain the MAC-IP route including the ESI sent by the second PE;

[0052] Specifically, according to the description of step 210, after the first PE obtains the ESI, it obtains the MAC-IP route sent by the second PE that includes the ESI.

[0053] In this embodiment of the application, the MAC-IP route synchronized by the PE includes the user-side MAC, ESI, and migration sequence number.

[0054] Step 230: Based on the MAC-IP route, obtain the first MAC entry generated corresponding to the MAC-IP route;

[0055] Specifically, according to the description of step 220, after the first PE obtains the MAC-IP route including the ESI, it retrieves the first MAC entry generated using the MAC-IP route from the local MAC table.

[0056] Optionally, the first MAC entry includes a status field that stores a first value indicating that the first MAC entry is a MAC entry generated via MAC-IP routing. The first value may be a label, such as EVPN.

[0057] Step 240: Convert the first MAC entry into the second MAC entry.

[0058] Specifically, according to the description of step 230, after the first PE obtains the first MAC table entry, it updates the status field of the first MAC table entry to obtain the second MAC table entry.

[0059] Optionally, the specific process of the first PE converting the first MAC entry into the second MAC entry is as follows: the first PE updates the first value stored in the status field to the second value to obtain the second MAC entry; the first PE starts the aging timer for the second MAC entry.

[0060] Optionally, the second MAC entry includes a status field that stores a second value indicating that the second MAC entry is a MAC entry generated through traffic learning. The second value may be specifically a label, such as "dynamic".

[0061] Therefore, using the communication method provided in this application, when the second PE fails, the ESI is obtained from the EAD per ES route sent by the second PE. The ESI is the identifier of the same ES associated with the first PE and the second PE. According to the ESI, the first PE obtains the MAC-IP route including the ESI sent by the second PE. According to the MAC-IP route, the first PE obtains the first MAC entry generated corresponding to the MAC-IP route. The first PE converts the first MAC entry into the second MAC entry.

[0062] Therefore, there is no need to start an additional timer to delete synchronously learned MAC entries. When MAC entries age normally, since the EAD per ES route has not been revoked, PE2 and PE3 receive the MAC-IP route revocation and can delete the MAC entries normally, avoiding the long-term occupation of MAC table resources by useless MAC entries. When handling EAD per ES route revocation, the MAC-IP routes synchronized by PEs associated with the same ES are relearned as local MAC entries. Even if the number of user-side MACs is huge and the PE's learning speed of user-side MACs is slow, there will be no packet loss due to MAC entry deletion. This solves the problem in the existing draft's delayed deletion scheme where, when the number of user-side MACs is huge, PE cannot guarantee that it will learn all user-side MACs within the time limit and there are residual user-side MACs in PE that are waiting to age.

[0063] Optionally, in this embodiment of the application, the first PE may also revoke any routes sent by the second PE other than the MAC-IP routes involved in the foregoing steps.

[0064] Specifically, the first PE obtains routes sent by the second PE other than the MAC-IP routes that include the aforementioned ESI. Other routes include, but are not limited to, MAC-IP routes that do not include ESI, EAD per EVI routes, EAD per ES routes, etc.

[0065] The first PE deletes the third MAC entries generated for other routes.

[0066] It should be noted that the route cancellation process for the third PE can refer to the handling method of PE3 provided in the draft draft-eastlake-bess-enhance-evpn-all-active-08 in the background technology, and will not be repeated here.

[0067] Please see Figure 1 The network shown is in Figure 1 In this configuration, PE1, PE2, and PE3 form an EVPN point-to-multipoint network, providing Layer 2 communication between CE1 and CE2. The forwarding plane between PE1, PE2, and PE3 can use VXLAN encapsulation, MPLS encapsulation, or SRv6 encapsulation. CE1 accesses the EVPN via an ES dual-homed network, identified as ESI, while CE2 accesses the EVPN in a single-homed manner.

[0068] PE1 and PE2 are both associated with the same Elasticsearch Entity (ES) and advertise EAD per EVI and EAD per ES routes to other PEs that have established BGP neighbor relationships. PE2 learns CE1's MAC address (MAC1) and synchronizes it with PE1 and PE3 via MAC-IP routes. After receiving the EAD per EVI and EAD per ES routes advertised by PE1 and PE2, as well as the MAC-IP route advertised by PE1, PE3 determines that CE1's ES access mode for the EVPN is multi-active ES. Therefore, the outgoing direction of the MAC entry in the locally generated MAC1 table is the equivalent path pointing to PE1 and PE2.

[0069] After receiving the EAD per EVI route, EAD per ES route, and MAC-IP route advertised by PE2, PE1 generates a MAC entry for MAC1 locally. The outgoing direction of the MAC entry points to the AC connected to CE1. At the same time, PE1 also advertises the MAC-IP route for MAC1.

[0070] After PE3 receives the MAC-IP route advertised by PE1, it determines that it is the same as the MAC-IP route advertised by PE2. PE3 will preferentially use one of the MAC-IP routes. Similarly, PE2 will also receive the MAC-IP route advertised by PE1. However, for PE2, since the MAC-IP route advertised by PE1 is the same as its own advertised MAC-IP route, PE2 will preferentially use the MAC1 route it learned, and will no longer process the MAC-IP route advertised by PE1.

[0071] When PE2 fails, all BGP neighbors between PE2 and PE1 and PE3 are lost. PE1 and PE3 will revoke all routes advertised by PE1. PE1 will retrieve the EAD per ES routes previously sent by PE2 from its local machine and obtain the ESI from the EAD per ES routes.

[0072] After obtaining the ESI, PE1 retrieves the MAC-IP route sent by PE2 that includes the ESI. From its local MAC table, PE1 obtains the first MAC entry generated using the MAC-IP route that includes the ESI. The first MAC entry includes a status field that stores the EVPN and indicates that the first MAC entry was generated via the MAC-IP route.

[0073] After PE1 obtains the first MAC entry, it updates the EVPN stored in the status field to "dynamic" to indicate that the second MAC entry is a MAC entry generated through traffic learning. PE1 then obtains the second MAC entry and starts the aging timer for it.

[0074] Based on the same inventive concept, embodiments of this application also provide a communication device corresponding to the communication method. See also Figure 3 , Figure 3 The communication device provided in this application embodiment is applied to a first PE, which is located within an EVPN network. The EVPN network also includes a second PE. The CE establishes connections with the first PE and the second PE respectively through different links. The first PE and the second PE form a multi-homed group. The device includes:

[0075] The first acquisition unit 310 is used to acquire the ESI from the EAD per ES route sent by the second PE when the second PE fails, wherein the ESI is the identifier of the same ES associated with the first PE and the second PE.

[0076] The second acquisition unit 320 is used to acquire the MAC-IP route including the ESI sent by the second PE according to the ESI;

[0077] The third acquisition unit 330 is used to acquire the first MAC entry generated corresponding to the MAC-IP route according to the MAC-IP route;

[0078] The conversion unit 340 is used to convert the first MAC entry into a second MAC entry.

[0079] Optionally, the first MAC entry includes a status field, which stores a first value, which indicates that the first MAC entry is a MAC entry generated through MAC-IP routing.

[0080] The second MAC entry includes the status field, which stores a second value indicating that the second MAC entry is a MAC entry generated through traffic learning.

[0081] Optionally, the conversion unit 340 is specifically used to update the first value stored in the status field to the second value to obtain the second MAC entry;

[0082] Start the aging timer for the second MAC entry.

[0083] Optionally, the apparatus further includes: a fourth acquisition unit (not shown in the figure), used to acquire routes other than the MAC-IP route sent by the second PE;

[0084] The deletion unit (not shown in the figure) is used to delete the third MAC entries generated for the other routes.

[0085] Therefore, using the communication device provided in this application, when the second PE fails, the ESI is obtained from the EAD per ES route sent by the second PE. The ESI is the identifier of the same ES associated with the first PE and the second PE. According to the ESI, the first PE obtains the MAC-IP route including the ESI sent by the second PE. According to the MAC-IP route, the first PE obtains the first MAC entry generated corresponding to the MAC-IP route. The first PE converts the first MAC entry into the second MAC entry.

[0086] Therefore, there is no need to start an additional timer to delete synchronously learned MAC entries. When MAC entries age normally, since the EAD per ES route has not been revoked, PE2 and PE3 receive the MAC-IP route revocation and can delete the MAC entries normally, avoiding the long-term occupation of MAC table resources by useless MAC entries. When handling EAD per ES route revocation, the MAC-IP routes synchronized by PEs associated with the same ES are relearned as local MAC entries. Even if the number of user-side MACs is huge and the PE's learning speed of user-side MACs is slow, there will be no packet loss due to MAC entry deletion. This solves the problem in the existing draft's delayed deletion scheme where, when the number of user-side MACs is huge, PE cannot guarantee that it will learn all user-side MACs within the time limit and there are residual user-side MACs in PE that are waiting to age.

[0087] Based on the same inventive concept, embodiments of this application also provide a network device, such as... Figure 4 As shown, the system includes a processor 410, a transceiver 420, and a machine-readable storage medium 430. The machine-readable storage medium 430 stores machine-executable instructions that can be executed by the processor 410. The processor 410 is prompted by the machine-executable instructions to execute the communication method provided in the embodiments of this application. (The foregoing...) Figure 3 The path protection device shown can be adopted as follows: Figure 4 The hardware structure of the network device shown is implemented.

[0088] The aforementioned computer-readable storage medium 430 may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the computer-readable storage medium 430 may also be at least one storage device located remotely from the aforementioned processor 410.

[0089] The processor 410 mentioned above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0090] In this embodiment of the application, the processor 410 reads the machine-executable instructions stored in the machine-readable storage medium 430, and is prompted by the machine-executable instructions to enable the processor 410 itself and the transceiver 420 to execute the communication method described in the foregoing embodiment of the application.

[0091] In addition, this application provides a machine-readable storage medium 430 that stores machine-executable instructions. When called and executed by the processor 410, the machine-executable instructions cause the processor 410 itself and the transceiver 420 to execute the communication method described in the aforementioned application.

[0092] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0093] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0094] For the embodiments of communication devices and machine-readable storage media, since the methods involved are basically similar to those of the aforementioned method embodiments, the description is relatively simple, and relevant details can be found in the descriptions of the method embodiments.

[0095] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A communication method, characterized in that, The method is applied to a first PE, which is located within an EVPN network. The EVPN network also includes a second PE. The PE establishes connections with the first PE and the second PE respectively through different links. The first PE and the second PE form a multi-homed group. The method includes: When the second PE fails, the ESI is obtained from the EAD per ES route sent by the second PE. The ESI is the identifier of the same ES associated with the first PE and the second PE. Based on the ESI, obtain the MAC-IP route including the ESI sent by the second PE; Based on the MAC-IP route, obtain the first MAC entry generated corresponding to the MAC-IP route; Convert the first MAC entry to the second MAC entry.

2. The method according to claim 1, characterized in that, The first MAC entry includes a status field, which stores a first value, which indicates that the first MAC entry is a MAC entry generated through MAC-IP routing. The second MAC entry includes the status field, which stores a second value indicating that the second MAC entry is a MAC entry generated through traffic learning.

3. The method according to claim 2, characterized in that, The step of converting the first MAC entry into the second MAC entry specifically includes: The first value stored in the status field is updated to the second value to obtain the second MAC table entry; Start the aging timer for the second MAC entry.

4. The method according to claim 1, characterized in that, The method further includes: Obtain routes other than the MAC-IP route sent by the second PE; Delete the third MAC entry generated for the other routes.

5. A communication device, characterized in that, The device is applied to a first PE, which is located within an EVPN network. The EVPN network also includes a second PE. The PE establishes connections with the first PE and the second PE via different links, respectively. The first PE and the second PE form a multi-homed group. The device includes: The first acquisition unit is used to acquire the ESI from the EAD per ES route sent by the second PE when the second PE fails, wherein the ESI is the identifier of the same ES associated with the first PE and the second PE; The second acquisition unit is used to acquire the MAC-IP route including the ESI sent by the second PE according to the ESI; The third acquisition unit is used to acquire the first MAC entry generated corresponding to the MAC-IP route based on the MAC-IP route; A conversion unit is used to convert the first MAC entry into a second MAC entry.

6. The apparatus according to claim 5, characterized in that, The first MAC entry includes a status field, which stores a first value, which indicates that the first MAC entry is a MAC entry generated through MAC-IP routing. The second MAC entry includes the status field, which stores a second value indicating that the second MAC entry is a MAC entry generated through traffic learning.

7. The apparatus according to claim 6, characterized in that, The conversion unit is specifically used to update the first value stored in the status field to the second value to obtain the second MAC entry; Start the aging timer for the second MAC entry.

8. The apparatus according to claim 5, characterized in that, The device further includes: The fourth acquisition unit is used to acquire routes other than the MAC-IP route sent by the second PE; The deletion unit is used to delete the third MAC entries generated for the other routes.

Images