CN-115643208-B - Communication method and device
Abstract
The application provides a communication method and a communication device, wherein the method is applied to a first PE, the first PE is located in an EVPN networking, the EVPN networking further comprises a second PE, a CE is respectively connected with the first PE and the second PE through different links, the first PE and the second PE form a multi-homing group, when the second PE fails, ESI is obtained from EAD per ES routes sent by the second PE, the ESI is identification of the same ES associated with the first PE and the second PE, MAC-IP routes including the ESI sent by the second PE are obtained according to the ESI, a first MAC table item correspondingly generated by the MAC-IP routes is obtained according to the MAC-IP routes, and the first MAC table item is converted into a second MAC table item.
Inventors
- XIE KEYANG
Assignees
- 新华三技术有限公司
Dates
- Publication Date
- 20260505
- Application Date
- 20220929
Claims (8)
- 1. A communication method, wherein the method is applied to a first PE, the first PE is in an EVPN network, the EVPN network further includes a second PE, the CE establishes a connection with the first PE and the second PE through different links, and the first PE and the second PE form a multi-homing group, the method includes: when the second PE fails, acquiring ESI from an EAD per ES route sent by the second PE, wherein the ESI is the identifier of the same ES associated with the first PE and the second PE; Acquiring an MAC-IP route which is sent by the second PE and comprises the ESI according to the ESI; Acquiring a first MAC table item correspondingly generated by the MAC-IP route according to the MAC-IP route; And converting the first MAC table entry into a second MAC table entry.
- 2. The method of claim 1, wherein the first MAC entry comprises a status field, the status field storing a first value indicating that the first MAC entry is a MAC entry generated by MAC-IP routing; the second MAC entry includes the status field, and the status field stores a second value, where the second value is used to indicate that the second MAC entry is a MAC entry generated by traffic learning.
- 3. The method according to claim 2, wherein said converting the first MAC entry into the second MAC entry specifically comprises: updating the first value stored in the status field to the second value to obtain the second MAC table item; and starting the aging timing of the second MAC table item.
- 4. The method according to claim 1, wherein the method further comprises: Acquiring other routes except the MAC-IP route sent by the second PE; and deleting the third MAC table entry correspondingly generated by the other routes.
- 5. A communication device, wherein the device is applied to a first PE, the first PE is in an EVPN network, the EVPN network further includes a second PE, the CE establishes a connection with the first PE and the second PE through different links, and the first PE and the second PE form a multi-homing group, the device includes: The first obtaining unit is configured to obtain, when the second PE fails, an ESI from an EAD per ES route sent by the second PE, where the ESI is an identifier of the same ES associated with the first PE and the second PE; A second obtaining unit, configured to obtain, according to the ESI, a MAC-IP route including the ESI sent by the second PE; A third obtaining unit, configured to obtain, according to the MAC-IP route, a first MAC entry generated corresponding to the MAC-IP route; and the conversion unit is used for converting the first MAC table entry into a second MAC table entry.
- 6. The apparatus of claim 5, wherein the first MAC entry comprises a status field, the status field storing a first value indicating that the first MAC entry is a MAC entry generated by MAC-IP routing; the second MAC entry includes the status field, and the status field stores a second value, where the second value is used to indicate that the second MAC entry is a MAC entry generated by traffic learning.
- 7. The apparatus of claim 6, wherein the converting unit is specifically configured to update the first value stored in the status field to the second value to obtain the second MAC entry; and starting the aging timing of the second MAC table item.
- 8. The apparatus of claim 5, wherein the apparatus further comprises: a fourth obtaining unit, configured to obtain other routes sent by the second PE, except for the MAC-IP route; and the deleting unit is used for deleting the third MAC table item correspondingly generated by the other routes.
Description
Communication method and device Technical Field The present application relates to the field of communications technologies, and in particular, to a communications method and apparatus. Background In the existing virtual private lan service (english: virtualPrivateLanService, abbreviated as VPLS) network, RFC7432 proposes a solution for fast convergence of unicast traffic, which can implement fast convergence of unicast traffic by revocation of the EAD per ES route. As shown in fig. 1, fig. 1 is a schematic diagram of a conventional VPLS network structure. In fig. 1, an ethernet virtual private network (english: ethernet Virtual Private Network, abbreviated: EVPN) is formed between service Provider Edge (english: provider Edge, abbreviated: PE) 1, PE2, and PE3 to point-to-multipoint networking, to provide two-layer interworking between Customer Edge (english: customer Edge, abbreviated: CE) 1, CE 2. The forwarding plane between PE1, PE2 and PE3 may employ virtual extended local area network (english: virtual Extensible Local Area Network, abbreviated: VXLAN) encapsulation, multiprotocol label switching (english: multi-Protocol Label Switching, abbreviated: MPLS) encapsulation, or IPv 6-based segment routing (english: segment Routing IPv6, abbreviated: SRv 6) encapsulation. CE1 is connected to EVPN through Ethernet segment (English: ETHERNET SEGMENT, short: ES). After learning the access local user side media access Control Address (english: MEDIA ACCESS Control Address, abbreviated: MAC) (e.g., virtual machine MAC of access PE, ethernet switch MAC, virtual machine MAC of access PE through ethernet switch, etc.), the PE synchronizes the user side MAC to other PEs of the established border gateway protocol (english: border Gateway Protocol, abbreviated: BGP) neighbors by the MAC-IP Route (also called MAC-IP publishing Route) through EVPN. For example, PE2 learns from the AC connected to CE1 to MAC1 of CE1 and synchronizes to PE1 and PE3. And after receiving the MAC-IP route issued by the PE2, the PE3 generates an MAC table entry of the MAC 1. If the ES mode of the CE1 accessing the EVPN is single active ES, the output direction of the MAC table entry is the main and standby paths pointing to PE1 and PE2, and if the ES mode of the CE1 accessing the EVPN is multi-active ES, the output direction of the MAC table entry is the equivalent path pointing to PE1 and PE 2. Similarly, on PE1, the exit direction of the MAC entry points to the AC connected to CE 1. When the link between the PE2 and the CE1 fails, on the PE3, if the ES mode is a multi-active ES, the PE3 needs to update the output direction of the MAC table from the equivalent path pointing to the PE1 and the PE2 to the single path pointing to the PE2, and if the ES mode is a single-active ES, the PE3 needs to update the output direction of the MAC table from the main path pointing to the PE1 and the PE2 to the standby path. Because the number of the user side MAC learned by the PE is huge, if the other PE finishes updating all the synchronous MAC-IP routes, a large number of flow messages can be lost due to wrong direction. Thus, RFC7432 proposes a solution for fast convergence of unicast traffic. That is, when a PE associates an ES, an EAD per ES route of EVPN is issued, which includes an Ethernet segment identifier (English: ETHERNET SEGMENT IDENTIFIER, ESI for short) and the address of the PE issuing the route. Meanwhile, the MAC-IP route issued by the PE also comprises the ESI and the user side MAC of the ES indicated by the access ESI. For example, after receiving the MAC-IP route issued by PE2 and the EAD per ES route issued by PE1 and PE2, the output direction of the MAC entry is different according to the ES mode. When the link between PE2 and CE1 fails, the EAD per ES route issued by PE2 is withdrawn and all traffic including MAC1 can no longer be forwarded to PE 2. Therefore, only one route is withdrawn, and the output direction of the MAC table entry can be modified. But the unicast traffic fast convergence scheme described above exposes a problem for node failures. When PE2 restarts, BGP neighbors between PE2 and PE1, PE3 are all disconnected. All routes published on PE1, PE3, equivalent to PE2, are not reachable (including EAD per ES route and MAC-IP route). Therefore, in PE3, the output direction of the MAC table entry is updated after the EAD per ES route is withdrawn, and the corresponding MAC table entry is deleted after the MAC-IP route is withdrawn. Similarly, the MAC entry is deleted in PE1, which results in changing unicast traffic to unknown unicast traffic in PE1 and PE 3. And switching the traffic to PE1 until the CE1 senses the PE2 fault, and after the PE1 relearns the user side MAC, issuing the traffic to the PE3, and changing the traffic to the known unicast traffic again. In practical applications, in order to prevent excessive traffic of Broadcast, unknown Unicast and multicast (in english: broadcast, unknown Unicast, multicast