WO2022227940A1 - Df选举模式切换方法、信息发送方法、pe设备和介质 - Google Patents

Df选举模式切换方法、信息发送方法、pe设备和介质 Download PDF

Info

Publication number
WO2022227940A1
WO2022227940A1 PCT/CN2022/082171 CN2022082171W WO2022227940A1 WO 2022227940 A1 WO2022227940 A1 WO 2022227940A1 CN 2022082171 W CN2022082171 W CN 2022082171W WO 2022227940 A1 WO2022227940 A1 WO 2022227940A1
Authority
WO
WIPO (PCT)
Prior art keywords
route
sub
interface
mode
advertised
Prior art date
Application number
PCT/CN2022/082171
Other languages
English (en)
French (fr)
Inventor
王玉保
陈然
戴春宁
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2022227940A1 publication Critical patent/WO2022227940A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H04L12/4641—Virtual LANs, VLANs, e.g. virtual private networks [VPN]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/26—Route discovery packet
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/28—Routing or path finding of packets in data switching networks using route fault recovery
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/50—Routing or path finding of packets in data switching networks using label swapping, e.g. multi-protocol label switch [MPLS]

Definitions

  • the present disclosure relates to the field of communication technologies, and in particular to a method for switching a Designated Forwarder (DF, Designated Forwarder) election mode in an Ethernet Virtual Private Network (EVPN, Ethernet Virtual Private Network) technology, a method for sending information, and a provider edge (PE, Provider Edge) devices and computer-readable media.
  • DF Designated Forwarder
  • EVPN Ethernet Virtual Private Network
  • PE provider edge
  • the DF election (DF Election) mode (hereinafter referred to as AC-DF mode, and the opposite is called non-AC-DF mode) affected by the sub-interface access circuit (AC, Attachment Circuit) is defined.
  • AC-DF mode the DF election (DF Election) mode
  • AC, Attachment Circuit the sub-interface access circuit
  • the purpose is to make the main interface of the Ethernet Segment Identifier (ESI, Ethernet Segment Identifier) not invalid, and the sub-interface identified by ESI and EVPN Instance (EVI, EVPN Instance) (represented by ⁇ ESI, EVI> later)
  • ESI Ethernet Segment Identifier
  • EVI, EVPN Instance the sub-interface identified by ESI and EVPN Instance (EVI, EVPN Instance) (represented by ⁇ ESI, EVI> later)
  • the interface fails, the failed sub-interface is excluded from the DF election of the ⁇ ESI, EVI>, so as to avoid the abnormal DF election
  • AC-DF mode relies on EVI-based Ethernet Auto-Discovery per EVI (Ethernet Auto-Discovery per EVI, RT-1 per EVI) routes. If an EVI-based Ethernet auto-discovery route is not published on an ⁇ ESI, EVI>, then The DF election on the ⁇ ESI, EVI> will be abnormal, and no matter whether there is a faulty sub-interface of the ⁇ ESI, EVI> on the PE node, problems will be caused due to the abnormal DF election.
  • EVI-based Ethernet Auto-Discovery per EVI Ethernet Auto-Discovery per EVI, RT-1 per EVI
  • EVPN scenarios for example, Provider Backbone Bridge (PBB, Provider Backbone Bridge) EVPN scenarios
  • PBB Provider Backbone Bridge
  • EVPN scenarios do not use EVI-based Ethernet to automatically discover routes. It is a waste to add and publish corresponding EVI-based Ethernet auto-discovery routes for all ESI sub-interfaces, which is a waste for scenarios where EVI-based Ethernet auto-discovery routes are not used originally.
  • the present disclosure provides a DF election mode switching method, an information sending method, a PE device and a computer-readable medium.
  • the present disclosure provides a method for switching a DF election mode, which is applied to a first PE device, where both the first PE device and the second PE device are on a first Ethernet segment (ES, Ethernet Segment), and the first PE device is provided with a The first sub-interface, the second PE device is provided with a second sub-interface corresponding to the first sub-interface, the first sub-interface and the second sub-interface are both bound to the first EVI, and the first sub-interface and the second sub-interface are The interfaces are all sub-interfaces on the first ES, and the method includes: in response to receiving the first route advertisement message advertised by the second PE device, switching the DF election mode corresponding to the first sub-interface to the first mode, the first mode A DF election mode that does not participate in the Ethernet segment route advertised by the second PE device for the first ES; and in response to receiving the first route revocation message advertised by the second PE device, electing the DF corresponding to the first sub-interface The mode switches to the second
  • the second mode is the DF election mode in which the Ethernet segment route advertised by the second PE device for the first ES participates.
  • the first route is the state of the second PE device for the second sub-interface.
  • the first route advertisement message is a route advertisement message that carries the multi-protocol network layer reachability information MP_REACH_NLRI and is published by the second PE device for the state change of the second sub-interface when the second sub-interface fails, and the first route advertisement message
  • a route revocation message is a route revocation message that carries the multi-protocol network layer unreachable information MP_UNREACH_NLRI and is published by the second PE device for the state change of the second sub-interface when the second sub-interface returns to normal.
  • the present disclosure also provides an information sending method, which is applied to a second PE device. Both the second PE device and the first PE device are on the first ES, the first PE device is provided with a first sub-interface, and the second PE device is provided with a first sub-interface.
  • the method includes: in response to the failure of the second sub-interface, publishing a first route advertisement message; and in response to the second sub-interface returning to normal, publishing a first route revocation message, wherein the first route is that the second PE device is directed to the second sub-interface
  • the first route advertisement message is a route advertisement message that carries the multi-protocol network layer reachability information MP_REACH_NLRI and is published by the second PE device for the state change of the second sub-interface when the second sub-interface fails
  • the first route revocation message is a route revocation message that carries the multi-protocol network layer unreachable information MP_UNREACH_NLRI and is published by the second PE device for the state change of the second sub-interface when the
  • the present disclosure also provides a PE device, comprising: one or more processors; and a memory on which one or more programs are stored, when the one or more programs are executed by the one or more processors, The one or more processors are caused to implement the DF election mode switching method or information sending method according to the present disclosure.
  • the present disclosure also provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, implements the DF election mode switching method or the information sending method according to the present disclosure.
  • FIG. 1 is a schematic flowchart of a switching method for a DF election mode provided by the present disclosure
  • FIG. 2 is a schematic flowchart of some steps in the switching method of the DF election mode provided by the present disclosure
  • FIG. 3 is a schematic flowchart of some steps in the switching method of the DF election mode provided by the present disclosure
  • FIG. 4 is a schematic flowchart of some steps in the switching method of the DF election mode provided by the present disclosure
  • FIG. 5 is a schematic flowchart of some steps in the switching method of the DF election mode provided by the present disclosure
  • FIG. 6 is a schematic flowchart of some steps in the switching method of the DF election mode provided by the present disclosure
  • FIG. 7 is a schematic flowchart of a method for sending information provided by the present disclosure.
  • FIG. 11 is a schematic flowchart of some steps in the information sending method provided by the present disclosure.
  • FIG. 13 is a schematic diagram of a DF election mode switching application scenario provided by the present disclosure.
  • FIG. 14 is a schematic diagram of a PE device provided by the present disclosure.
  • 15 is a schematic diagram of a computer-storable medium provided by the present disclosure.
  • the basic network scenario of the embodiment of the present disclosure is as follows: the same customer edge (CE, Customer Edge) device accesses the first PE device and the second PE device in a dual-homing manner, the first PE device is provided with a first sub-interface, and the second PE device is provided with a first sub-interface.
  • the PE device is provided with a second sub-interface corresponding to the first sub-interface.
  • the switching operation of the DF election mode corresponding to the first sub-interface of the first PE device is triggered.
  • the present disclosure provides a method for switching the DF election mode, which is applied to the first PE device. As shown in FIG. 1 , the method includes the following steps S100 to S200.
  • step S100 in response to receiving the first route advertisement message advertised by the second PE device, the DF election mode corresponding to the first sub-interface is switched to the first mode.
  • the first mode is a DF election mode without the participation of the Ethernet segment route advertised by the second PE device for the first ES.
  • Both the first sub-interface and the second sub-interface are bound to the first EVI, and both the first sub-interface and the second sub-interface are sub-interfaces on the first ES.
  • the first route advertisement message is a route advertisement message that carries the multi-protocol network layer reachability information MP_REACH_NLRI and is published by the second PE device for the state change of the second sub-interface when the second sub-interface fails.
  • the first PE device After the first PE device receives the first route advertisement message, it will determine the first sub-interface corresponding to it according to the ESI information indicating the first ES and the EVI information indicating the first EVI included in the first route.
  • the Ethernet segment route advertised by the PE device for the first ES is excluded from the DF election process corresponding to the first sub-interface.
  • the first mode is adopted, and the DF election on the ⁇ ESI, EVI> corresponding to the first sub-interface will exclude the second sub-interface, thereby avoiding the ⁇ ESI, EVI> of the first sub-interface.
  • the DF election result of ESI, EVI> is abnormal.
  • the second sub-interface is excluded, that is, the Ethernet segment route advertised by the PE node where the second sub-interface is located (ie, the second PE device) is excluded.
  • the second PE device only needs to cancel the corresponding Ethernet segment route, and does not need to advertise the first route to each sub-interface. Announcement message. Therefore, according to the received first route advertisement message, the first PE device can know that the corresponding sub-interface of the corresponding ES on the second PE device is invalid but its main interface is not invalid.
  • the first PE device After receiving the first route advertisement message advertised by the second PE device, the first PE device switches the DF election mode on the corresponding ⁇ ESI, EVI> from the second mode to the first mode.
  • step S200 in response to receiving the first route revocation message advertised by the second PE device, the DF election mode corresponding to the first sub-interface is switched to the second mode.
  • the second mode is a DF election mode in which the Ethernet segment route advertised by the second PE device for the first ES participates.
  • the first route revocation message is a route revocation message that carries the multi-protocol network layer reachability information MP_UNREACH_NLRI and is published by the second PE device for the state change of the second sub-interface when the second sub-interface returns to normal.
  • the first PE device After receiving the first route revocation message, the first PE device determines the corresponding first sub-interface according to the ESI information indicating the first ES and the EVI information indicating the first EVI included in the first route, so that the second The PE device resumes and participates in the DF election process corresponding to the first sub-interface for the Ethernet segment route advertised by the first ES, just like before the failure of the second sub-interface. Therefore, the second mode is called "with the second PE". DF election mode in which the device participates in the Ethernet segment route advertised by the first ES".
  • each network element receives multiple first route advertisement messages. Because each first route revocation message can only revoke its corresponding first route advertisement message, the first PE device cannot directly cancel the DF corresponding to the first sub-interface when only one first route revocation message is received.
  • the election mode switches to the second mode, but should switch to the second mode after receiving the respective revocation messages corresponding to all the first route advertisement messages.
  • the first PE device After receiving the first route revocation message advertised by the second PE device, the first PE device switches the DF election mode on the corresponding ⁇ ESI, EVI> to the second mode before step S100.
  • the initial DF election mode on each ⁇ ESI,EVI> is the second mode (for example, non-AC-DF mode), no need to publish EVI-based Ethernet Auto-Discovery per EVI, RT-1 per EVI ) route; when there is a sub-interface failure on a certain ⁇ ESI, EVI>, the DF election mode on the ⁇ ESI, EVI> is switched to the first mode (for example, AC-DF mode).
  • the first mode for example, AC-DF mode.
  • only the ⁇ ESI, EVI> needs to be advertised with a route equivalent to the magnitude of the faulty sub-interface, and each ⁇ ESI, EVI> enters the first mode respectively.
  • correct results can be obtained by DF election in the first mode or the second mode, and the second mode has better performance because the number of routes to be advertised is reduced by an order of magnitude. good.
  • step S101 may be further included.
  • step S101 in response to receiving the first route advertisement message advertised by the second PE device and the state of the first sub-interface is normal, a route advertisement message indicating that the state of the first sub-interface is normal is published for the first sub-interface.
  • the first PE device does not need to publish a route advertisement message specifically for a specific sub-interface that is not faulty to its neighbors for the purpose of DF election, so that routes can be reduced by an order of magnitude. Announcement message.
  • the EVI-based Ethernet auto-discovery route was not originally published. Before switching to the first mode, the EVI-based Ethernet auto-discovery route was not used to obtain each network within the range of ⁇ ESI, EVI>. Subinterface state information of the element. Therefore, when receiving the first route advertisement message for the failure of the second sub-interface advertised by the second PE device, the first PE device needs to reissue the first PE to the EVPN neighbor within the range of the first route ⁇ ESI, EVI> Route advertisement information about the normal operation of the first sub-interface corresponding to the second sub-interface on the device. The route advertisement information is used to notify the neighbor that the state of the first sub-interface is normal.
  • all network elements in the ⁇ ESI, EVI> range can perform step S101, so each network element can know the node information of all the normal working sub-interfaces in the ⁇ ESI, EVI> range. , so that the node where the sub-interface that cannot work normally is located is excluded from the DF election process of the ⁇ ESI, EVI>.
  • first sub-interface and the second sub-interface belong to the same ⁇ ESI, EVI>, and the first route advertisement message and the first route withdrawal message are also directed to ⁇ ESI, EVI>. Specifically, in the example, it is all for ⁇ first ESI, first EVI>, so they can be indexed to each other through ⁇ ESI, EVI> information.
  • step S101 does not need to be performed, and each network element defaults to the node that already knows the first route advertisement message (ie, the second PE device ), the corresponding sub-interfaces of the ⁇ ESI, EVI> are valid on all other nodes, so that the nodes whose corresponding sub-interfaces have failed are excluded from the DF election process of the ⁇ ESI, EVI>.
  • excludede a node from the DF election process means to exclude the Ethernet segment route advertised by the node from the DF election process of the corresponding ⁇ ESI, EVI>, that is, to exclude it from being used by the DF election process. Possibility of election as DF.
  • step S102 is further included.
  • step S102 in response to receiving the Ethernet segment route carrying the first indication information issued by the second PE device, set the DF election mode on each ⁇ ESI, EVI> in the first ES to the second mode .
  • the first indication information is indication information advertised when the PE device can advertise and receive the first route.
  • a node that implements the solution of the present disclosure can confirm whether there is a node that does not implement the solution of the present disclosure in the network. Only when all PE devices on the same ES implement the solution of the present disclosure, the node that implements the solution of the present disclosure actually publishes the first route advertisement message to the outside world.
  • the first PE device receives the Ethernet segment route that carries the first indication information advertised by the second PE device, it will judge whether it has received the first route advertisement message, and it will determine whether the first route advertisement message has been received by the first ES device.
  • the DF election on ⁇ ESI, EVI> of a route advertisement message is set to the second mode, and the DF election on all ⁇ ESI, EVI>s that have received the first route advertisement message is set to the first mode.
  • step S103 is further included.
  • step S103 the DF election mode corresponding to the first sub-interface is switched to the first mode only when all the received Ethernet segment routes of the first ES carry the first indication information.
  • the initial DF election mode on each ⁇ ESI,EVI> is the second mode (for example, non-AC-DF mode), no need to publish EVI-based Ethernet Auto-Discovery per EVI, RT-1 per EVI ) route; when there is a sub-interface failure on a certain ⁇ ESI, EVI>, the DF election mode on the ⁇ ESI, EVI> is switched to the first mode (for example, AC-DF mode).
  • the ⁇ ESI, EVI> only needs to be advertised on a route (for example, the first route) equivalent to the magnitude of the EVI-based Ethernet automatic discovery route, and each ⁇ ESI, EVI> enters the first mode respectively.
  • the second mode has better performance because the number of routes to be advertised is reduced by an order of magnitude. good.
  • step S104 is further included.
  • step S104 the DF election mode corresponding to the first sub-interface is switched to the second mode only when all the received Ethernet segment routes of the first ES carry the first indication information.
  • step S103 when the second sub-interface returns to normal, it is necessary to notify the DF election mode on the corresponding ⁇ ESI, EVI> to switch to the second mode before step S103 through the first route revocation message.
  • the Ethernet segment route advertised by the second PE device for the first ES is excluded from the DF election process corresponding to the first sub-interface.
  • the first route can be an EVI-based Ethernet automatic discovery route that additionally carries the second indication information, wherein the second indication information can be carried in the EVPN Layer 2 Attributes Extended Community attribute, and specifically, can be It is carried in a flag bit in the "Control Flags" field of this attribute.
  • the flag bit carrying the second indication information is called an AC failure flag bit.
  • the second PE device publishes an Ethernet auto-discovery route advertisement message in this format, in which the value of the AC failure flag is set to "1" to tell the receiving end PE device that it is the first route Announcement message.
  • the AC failure flag bit may not be limited to 1 bit, nor is it limited to be set to "1" to indicate that the AC fails, as long as the setting includes the meaning of AC failure, it is within the protection scope of the present disclosure.
  • the first PE device After receiving the first route advertisement message, the first PE device excludes the corresponding Ethernet segment route from the DF election process.
  • the route target membership (Route Target Constraints, RT Constraints) mechanism of the Border Gateway Protocol (BGP, Border Gateway Protocol) can be advertised to the neighbors including the first route target. membership) Network Layer Reachability Information (NLRI, Network Layer Reachability Information).
  • NLRI Network Layer Reachability Information
  • the EVPN route of the second PE device is not restricted when it is advertised, and will carry the route target to all neighbors. Limiting the ES range by routing targets can avoid occupying routing resources on unnecessary neighbors.
  • the routing target can use special types, such as ES inbound routing target (ES Import RT) and EVI routing target (EVI-RT), the purpose is to carry the routing target of the first ES and the first EVI in the first routing advertisement message at the same time.
  • ES Import RT ES inbound routing target
  • EVI-RT EVI routing target
  • the node receiving the first route advertisement message can distinguish the two routing targets, so that the import scope of the first route can be controlled on the PE device in the same ES, while other unrelated PE devices in the network do not Routes with these route targets will be accepted and imported only if both ES Import RT and EVI-RT match.
  • the route forwarding path there may be nodes running the RT Constraints mechanism (for example, a certain route reflector (RR, Route Reflector) node), which will restrict the EVPN routes advertised by the second PE device.
  • the route target membership NLRI containing the first route target is advertised to neighbors through the RT Constraints mechanism of BGP, so that the first route target A route is forwarded smoothly in the network.
  • step S105 may also be included.
  • step S105 in response to receiving the first route advertisement message with the route target defining the first ES issued by the second PE device, and the third PE device does not locally have an ES matching the route target, the third PE device ignores the first route advertisement message.
  • the second PE device and the third PE device have an EVPN neighbor relationship, and none of the sub-interfaces of the third PE device are on the first ES.
  • the first route advertisement message advertised by the second PE device contains a route target that defines the first ES, because none of the sub-interfaces of the third PE device are on the first ES, the third PE device does not locally exist and the route target matching ES, so the first route advertisement message is ignored.
  • the present disclosure also provides an information sending method, which is applied to the second PE device, as shown in FIG. 7 , and includes the following steps S300 to S400.
  • step S300 in response to the failure of the second sub-interface, a first route advertisement message is published.
  • Both the first sub-interface and the second sub-interface are bound to the first EVI, and both the first sub-interface and the second sub-interface are sub-interfaces on the first ES.
  • the neighbor network element When the main interface of the ESI does not fail but its sub-interface fails, the neighbor network element needs to be notified that the failed sub-interface (for example, the second sub-interface) is located in the DF election process on the corresponding sub-interface ⁇ ESI, EVI>.
  • the node ie, the node that publishes the first route advertisement message
  • the DF election result on ⁇ ESI, EVI> will be abnormal.
  • the first route advertisement message is a route advertisement message that carries the multi-protocol network layer reachability information MP_REACH_NLRI and is published by the second PE device for the state change of the second sub-interface when the second sub-interface fails.
  • the first route may be by adding a special flag to an existing route type, or by adding a new address family, or by adding a new route type. For example, use a new EVPN route type with a route type code greater than 13.
  • the present disclosure uses the multi-protocol network layer reachability information MP_REACH_NLRI instead of the multi-protocol network layer unreachable information MP_UNREACH_NLRI.
  • MP_REACH_NLRI is for route advertisement
  • MP_UNREACH_NLRI is for revocation of MP_REACH_NLRI. If MP_REACH_NLRI corresponding to MP_UNREACH_NLRI has not been received before, the route revocation message carrying MP_UNREACH_NLRI is invalid. The usual practice is to discard the route revocation message carrying the MP_UNREACH_NLRI.
  • the first route in the present disclosure is to advertise the "failure" of the sub-interface to neighbors as the content of the routing message, and the "failure" of the sub-interface and the "reachable” of MP_REACH_NLRI are not the same concept.
  • the second PE device uses a route advertisement message carrying MP_REACH_NLRI to announce the failure of the second sub-interface. After receiving the route advertisement message, the RR forwards the route advertisement message. to the first PE device.
  • step S400 in response to the second sub-interface returning to normal, a first route revocation message is published.
  • the first route uses the route revocation message carrying MP_UNREACH_NLRI to cancel the corresponding route advertisement message carrying MP_REACH_NLRI issued when the sub-interface "fails" in step S300.
  • the second PE advertises a route revocation message carrying MP_UNREACH_NLRI, which is used to revoke the previously advertised corresponding route advertisement message carrying MP_REACH_NLRI, and the RR receives it.
  • step S301 may also be included.
  • step S301 before publishing the first route advertisement message, the Ethernet segment route carrying the first indication information is published.
  • the first indication information is indication information advertised when the PE device can advertise and receive the first route.
  • the DF election extended community (Extended Community) attribute can be extended, and a flag bit indicating switching to AC-DF mode based on EVI can be extended in the "Bitmap" field, as an implementation of the first indication information to indicate the second
  • the PE device implements the disclosed scheme (ie, is capable of sending and receiving the first route).
  • the second PE device is said to execute the EVI-based AC-DF (referred to as AC-DF per EVI) mode on the first ES.
  • EVI-based AC-DF mode is: the initial state (that is, the state without any faults) each ⁇ ESI, EVI> on the ES adopts the second mode, and when an ⁇ ESI, EVI> fault is sensed After that, the first mode is separately enabled at the ⁇ ESI, EVI> granularity, and after the ⁇ ESI, EVI> fault is removed, the second mode is restored.
  • the EVI-based AC-DF mode refers to the overall situation of the DF election mode (ie, the first mode or the second mode) adopted by each ⁇ ESI, EVI> on an ES, and the first mode and the first mode
  • the second mode refers to the DF election mode on a specific ⁇ ESI, EVI> of the ES.
  • the first indication information also indicates that the overall situation of the DF election mode adopted by each ⁇ ESI, EVI> of the corresponding ESI is the EVI-based AC-DF mode.
  • the first indication information indicates that the second PE device implements the solution of the present disclosure, or the first indication information indicates that the second PE device can send and receive the first route, or the first indication information indicates that each of the first ESs where the second PE device is located.
  • EVI EVI>
  • the overall situation of the DF election mode adopted is the AC-DF mode based on EVI, which is a different expression of the same meaning.
  • Ethernet segment routes carry the first indication information
  • each PE node on the corresponding ES implements the solution of the present disclosure.
  • the first route is negotiated at the ES level to support the first route.
  • the DF election mode is negotiated as the EVI-based AC-DF mode at the ES granularity, or the overall situation of the DF election mode adopted by each ⁇ ESI, EVI> of the ES is negotiated as the EVI-based AC-DF mode.
  • DF mode these terms are different expressions of the same meaning.
  • the first indication information may be set simultaneously with the AC-DF capability (AC-DF Capability) flag.
  • the first indication information and the AC-DF capability flag in the DF election mode flag can be set simultaneously.
  • the AC-DF capability flag When the first indication information and the AC-DF capability flag in the Bitmap field are set at the same time, if all Ethernet segment routes of the ES carry the first indication information (it means that there is no node in the ES that does not implement the solution of the present disclosure) ), the AC-DF capability flag is automatically invalid; otherwise, the AC-DF capability flag is still valid. In this way, it can be ensured that when there is a node that does not support the first indication information in the ES, the DF can still be elected according to the method indicated by the AC-DF capability flag, so that the nodes that support AC-DF but do not implement the solution of the present disclosure can better play the role of DF electoral capacity.
  • step S302 in response to the failure of the second sub-interface, the following step S302 may be included.
  • step S302 the first route advertisement message is distributed only when the Ethernet segment routes advertised by all PE devices carry the first indication information.
  • the first indication information is indication information advertised when the PE device can advertise and receive the first route.
  • the first routing message may not add a new routing type, but still use the existing EVI-based Ethernet automatic discovery routing message to notify the first PE device of the state change of the second sub-interface.
  • the advantage of this is that it is easier to be compatible with existing networks (eg RR nodes therein).
  • a node that implements the solution of the present disclosure can confirm whether there is a node that does not implement the solution of the present disclosure in the network. Only when all PE devices on the same ES implement the solution of the present disclosure, the node that implements the solution of the present disclosure actually publishes the first route advertisement message to the outside world. In this way, when there is a node in the network that does not implement the solution of the present disclosure (ie, does not carry the first indication information in the Ethernet segment route), all Ethernet segment routes of the corresponding ES may not carry the first indication information According to the rules at the time of processing, it is ensured that the results will not be worsened when the old and new nodes are mixed in the network.
  • step S303 in response to the failure of the second sub-interface, the following step S303 may also be included.
  • step S303 in the first route advertisement message, the Multiprotocol Label Switching Protocol Label (MPLS Label, Multiprotocol Label Switching Protocol Label) field is set as a reserved value, and the Ethernet Tag ID (Ethernet Tag ID) of the first route advertisement message ) field is set as the I component service identifier (I-SID, I-Component Service ID) value corresponding to the first EVI, and the route of the first backbone component (B-Component, Backbone Component) is carried in the first route advertisement message The target, wherein the first backbone component is a backbone component bound to the first EVI.
  • MPLS Label, Multiprotocol Label Switching Protocol Label MPLS Label, Multiprotocol Label Switching Protocol Label
  • Ethernet Tag ID Ethernet Tag ID
  • I-SID I-Component Service ID
  • step S301 in order to be different from the conventional settings in MP_REACH_NLRI, in the EVI-based Ethernet automatic route discovery, special settings are made to MP_REACH_NLRI, including setting the MPLS Label field to a reserved value that does not conflict with existing protocols , such as 0, and set the Ethernet Tag ID field to the I-SID value corresponding to the first EVI.
  • the first route advertisement message carries second indication information for identifying the route as the first route.
  • the first route may be a newly added route type (in this case, the value of the route type is the second indication information), or may be a route specially set on the EVI-based Ethernet automatic discovery route.
  • the first route may use the L2VPN/EVPN address family, whose NLRI type is 1, and add an AC failure flag as the second indication information in the Control Flags field of the extended community attribute of the EVPN Layer 2 attribute.
  • the receiving side can learn that the second sub-interface of the second PE device is invalid according to the AC failure flag in the extended community attribute, and exclude the Ethernet segment route advertised by the second PE device for the first ES from the corresponding first sub-interface. outside the DF election process.
  • step S304 in response to the failure of the second sub-interface, the following step S304 may also be included.
  • step S304 the first routing target corresponding to the first ES is carried in the first routing advertisement message.
  • the EVPN route of the second PE device is not restricted when it is advertised, and will carry the route target to all neighbors. Limiting the ES range by routing targets can avoid occupying routing resources on unnecessary neighbors.
  • the first route may be a newly added route type, or may be a route specially set on the EVI-based Ethernet automatic discovery route.
  • the filling method of the MP_REACH_NLRI field is the same as that of the EVI-based Ethernet auto-discovery route, but its route target adopts a special type, such as ES Import RT and EVI-RT.
  • the purpose is to carry the route of the first ES in the first route advertisement message at the same time.
  • the node receiving the first route advertisement message can distinguish the two route destinations, so that the import range of the first route can be controlled on the PE device in the same ES, and other unrelated items in the network can be controlled.
  • the PE devices with these routes will not accept routes with these route targets.
  • the presence of ES Import RT and/or EVI-RT is the second indication information.
  • the first route target may be visible to the RT Constraints mechanism.
  • the RT Constraints mechanism may run, which will restrict the EVPN routes advertised by the second PE device.
  • the NLRI of the route target membership including the first route target is advertised to neighbors through the RT Constraints mechanism of BGP.
  • the first routing target is visible to the RT Constraints mechanism, that is, the RT Constraints mechanism checks the first routing target, and if the first routing target cannot match the routing target membership, the route with the first routing target will not be published.
  • the first route target may be used to cause PE nodes not on the first ES to filter the first route advertisement message.
  • the first route advertisement message carries both the route target of the first ES and the route target of the first EVI. Since the PE node (eg, the third PE device) is not on the first ES, the route target of the first ES cannot be matched and the first route advertisement message is ignored.
  • the PE node eg, the third PE device
  • step S305 may also be included.
  • step S305 the first route advertisement message carries the second route target corresponding to the first EVI, and the second route target is invisible to the RT Constraints mechanism.
  • the first routing advertisement message will be due to the same If the route target corresponding to the first ES does not match, it will not be forwarded to the third PE device, thus ensuring that in the network where RT Constraints are deployed, the first route advertisement message will only reach the PE node on the first ES.
  • the third PE device since the third PE device is configured with the first EVI, it will advertise the routing target membership NLRI for the routing target of the first EVI (ie, the second routing target), but will not publish the routing target ( That is, the first routing target membership NLRI, therefore, although the first routing advertisement message contains the second routing target invisible to the RT-Constraints mechanism, the first routing advertisement message will also be due to the first routing target. There is no match to the route target membership NLRI advertised by the third PE device and not advertised to the third PE.
  • PE1 and PE2 are the first PE device and the second PE device respectively (it is not limited that PE1 is the first PE device and PE2 is the second PE device.
  • the specific roles are The corresponding relationship depends on the actual situation)
  • PE3 is an exemplary other PE device (such as a third PE device)
  • R4 and R5 are other exemplary devices between the PE devices
  • the ESI of the ES where the CE device is located is ESI2
  • ESI2 is bound to the main interface IF_1, but services are bound to the sub-interfaces IF_1.1 and IF_1.2 of IF_1.
  • the sub-interfaces IF_1.1 and IF_1.2 are bound in the I component (ie, EVPN instance, EVI) as ACs, the first route is advertised in the I component, and the Ethernet auto-discovery route is not advertised.
  • the I component is the I component of the PBB EVPN.
  • the PBB EVPN also has a B component (ie, a backbone component (B-Component)).
  • B-Component backbone component
  • the example shows two EVPN instances (EVI), which are 100 and 200 respectively.
  • each EVI has a corresponding I component on each PE device, and the DF election must select a unique DF node for each ⁇ ESI, EVI>.
  • the DF election result is: PE1 is DF and PE2 is non-DF; for ⁇ ESI2, EVI200>, the DF election result is: PE1 is non-DF and PE2 is DF.
  • the second mode is used for the DF election of ESI2, that is, the current DF election result is not affected by the
  • the second indication information is that the EVI-based Ethernet auto-discovery route exists or not.
  • PE1 advertises the first route advertisement message to the sub-interface IF_1.1, and other PE devices on the same ES (take PEn as an example) can pass The first route advertisement message switches the DF election on the corresponding ⁇ ESI, EVI> from the second mode to the first mode.
  • PE1 When the sub-interface IF_1.1 on PE1 is restored, PE1 advertises the revocation message of the first route. When PEn receives the revocation message, it cancels the first route advertisement message previously received from PE1, and the ⁇ ESI, EVI > Switch from the first mode to the second mode on the DF election. If there are no other first route advertisement messages on the same ⁇ ESI, EVI>, all the Ethernet segment routes advertised by the PE devices for the ESI will participate in the DF election process for the ⁇ ESI, EVI>.
  • VXLAN virtual extensible local area network
  • SRv6, Segment Route IPv6 segment routing IPv6
  • MAC/IP Advertisement Route instead of using the MAC/IP address advertisement route (MAC/IP Advertisement Route), the remote MAC entry can be learned on the data plane to replace the role of the MAC/IP address advertisement route.
  • the first route advertisement message may only carry the route target of the B component but not the route target of the I component.
  • the Ethernet Tag ID field in the first route advertisement message can be Set to the corresponding I-SID to distinguish the first route advertisement messages for different I components.
  • the first route can be a new EVPN route type (when the second indication information is its route type), which is opposite to the timing of publishing and revoking the Ethernet auto-discovered route, that is, the latter
  • the release timing of the first route is the cancellation timing of the first route
  • the cancellation timing of the latter is the advertisement timing of the first route. Therefore, the first route is published when a fault occurs, and is cancelled when the fault is recovered.
  • the first route will not be advertised, that is, the first route will not occupy network resources under normal conditions. Only when there is a sub-interface failure on an ⁇ ESI, EVI>, the DF election needs to be switched to the first mode, and it is necessary to advertise the first route equivalent to the faulty sub-interface.
  • the present disclosure provides a PE device including: one or more processors 501 ; and a memory 502 on which one or more programs are stored, when the one or more programs are stored by one or more
  • the processor 501 executes, so that one or more processors 501 implement the switching method or the information sending method of the DF election mode according to various embodiments of the present disclosure.
  • I/O interfaces 503 may also be included, which are connected between the processor 501 and the memory 502 and are configured to implement information interaction between the processor 501 and the memory 502 .
  • the processor 501 may be a device with data processing capabilities, including but not limited to a central processing unit (CPU), etc.; the memory 502 may be a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically Such as SDRAM, DDR, etc.), read-only memory (ROM), electrified erasable programmable read-only memory (EEPROM), flash memory (FLASH); the I/O interface 503 is connected between the processor 501 and the memory 502, which can realize the processor 501 interacts with the information of the memory 502, which includes but is not limited to a data bus (Bus) and the like.
  • RAM random access memory
  • ROM read-only memory
  • EEPROM electrified erasable programmable read-only memory
  • FLASH flash memory
  • the I/O interface 503 is connected between the processor 501 and the memory 502, which can realize the processor 501 interacts with the information of the memory 502, which includes but is not limited to a data bus (Bus) and the like.
  • the processor 501, memory 502, and I/O interface 503 may be connected to each other through a bus 504 and, in turn, to other components of the computing device.
  • the present disclosure provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, implements the DF election mode switching method or information sending method according to various embodiments of the present disclosure.
  • Such software may be distributed on computer-readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media).
  • computer storage media includes both volatile and nonvolatile implemented in any method or technology for storage of information, such as computer readable instructions, data structures, program modules or other data flexible, removable and non-removable media.
  • Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cartridges, magnetic tape, magnetic disk storage or other magnetic storage devices, or may Any other medium used to store desired information and that can be accessed by a computer.
  • communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and can include any information delivery media, as is well known to those of ordinary skill in the art .

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本公开提供一种DF选举模式的切换方法、信息发送方法、PE设备和计算机可读介质。所述DF选举模式的切换方法包括：响应于接收到第二PE设备针对第二子接口失效而发布的第一路由通告消息，第一PE设备将DF选举模式切换到第一模式；以及响应于接收到第二PE设备针对第二子接口恢复正常而发布的第一路由撤销消息，第一PE设备将DF选举模式切换到第二模式。

Description

DF选举模式切换方法、信息发送方法、PE设备和介质 技术领域
本公开涉及通信技术领域，特别涉及以太网虚拟专用网络(EVPN，Ethernet Virtual Private Network)技术中的指定转发者(DF，Designated Forwarder)选举模式的切换方法、信息发送方法、提供商边缘(PE，Provider Edge)设备和计算机可读介质。
背景技术
在EVPN技术中，定义了受子接口接入电路(AC，Attachment Circuit)影响的DF选举(DF Election)模式(后面称为AC-DF模式，与之相对的则称为非AC-DF模式)，其目的是使得当以太网分段标识符(ESI，Ethernet Segment Identifier)的主接口没有失效，而由ESI和EVPN实例(EVI,EVPN Instance)(后面由<ESI,EVI>表示)标识的子接口失效时，在对该<ESI,EVI>的DF选举中，将失效子接口排除在外，从而避免该<ESI,EVI>的DF选举结果异常。AC-DF模式依赖于基于EVI的以太网自动发现(Ethernet Auto-Discovery per EVI，RT-1 per EVI)路由，如果某个<ESI,EVI>上没有发布基于EVI的以太网自动发现路由，则该<ESI,EVI>上的DF选举就会异常，而且无论是否有PE结点上存在该<ESI,EVI>的故障子接口，都会由于DF选举异常而导致问题。
存在一种机制，定义了DF选举扩展团体属性，用于协商DF选举模式是否为AC-DF模式，但是，该协商机制是基于ESI粒度的，而不是基于<ESI,EVI>粒度的，因此，一旦在ESI粒度上协商为AC-DF模式，则该ESI的所有<ESI,EVI>上都必须发布基于EVI的以太网自动发现路由，否则没有发布基于EVI的以太网自动发现路由的<ESI,EVI>上的DF选举结果就会异常。
但是，某些EVPN场景(比如，提供商骨干网桥(PBB，Provider Backbone Bridge)EVPN场景)并不使用基于EVI的以太网自动发现路由，如果仅仅为了进行受AC影响的DF选举而导致常态下就对所有 ESI的子接口都增加发布相应的基于EVI的以太网自动发现路由，对于原本不使用基于EVI的以太网自动发现路由的场景而言是一种浪费。
发明内容
本公开提供一种DF选举模式的切换方法、信息发送方法、PE设备和计算机可读介质。
本公开提供一种DF选举模式的切换方法，应用于第一PE设备，第一PE设备与第二PE设备均在第一以太网分段(ES，Ethernet Segment)上，第一PE设备设置有第一子接口，第二PE设备设置有与第一子接口相对应的第二子接口，第一子接口及第二子接口均绑定了第一EVI，且第一子接口及第二子接口均为第一ES上的子接口，方法包括：响应于接收到第二PE设备发布的第一路由通告消息，将与第一子接口对应的DF选举模式切换到第一模式，第一模式为没有第二PE设备为第一ES发布的以太网分段路由参与的DF选举模式；以及响应于接收到第二PE设备发布的第一路由撤销消息，将与第一子接口对应的DF选举模式切换到第二模式，第二模式为有第二PE设备为第一ES发布的以太网分段路由参与的DF选举模式，其中，第一路由为第二PE设备针对第二子接口的状态变化而发布的路由，第一路由通告消息是当第二子接口失效时由第二PE设备针对第二子接口的状态变化发布的携带多协议网络层可达信息MP_REACH_NLRI的路由通告消息，并且第一路由撤销消息是当第二子接口恢复正常时由第二PE设备针对第二子接口的状态变化发布的携带多协议网络层不可达信息MP_UNREACH_NLRI的路由撤销消息。
本公开还提供一种信息发送方法，应用于第二PE设备，第二PE设备与第一PE设备均在第一ES上，第一PE设备设置有第一子接口，第二PE设备设置有与第一子接口相对应的第二子接口，第一子接口及第二子接口均绑定了第一EVI，且第一子接口及第二子接口均为第一ES上的子接口，方法包括：响应于第二子接口失效，发布第一路由通告消息；以及响应于第二子接口恢复正常，发布第一路由撤销消 息，其中，第一路由为第二PE设备针对第二子接口的状态变化而发布的路由，第一路由通告消息是当第二子接口失效时由第二PE设备针对第二子接口的状态变化发布的携带多协议网络层可达信息MP_REACH_NLRI的路由通告消息，并且第一路由撤销消息是当第二子接口恢复正常时由第二PE设备针对第二子接口的状态变化发布的携带多协议网络层不可达信息MP_UNREACH_NLRI的路由撤销消息。
本公开还提供一种PE设备，包括：一个或多个处理器；以及存储器，其上存储有一个或多个程序，当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现根据本公开的DF选举模式的切换方法或信息发送方法。
本公开还提供一种计算机可读介质，其上存储有计算机程序，所述程序被处理器执行时实现根据本公开的DF选举模式的切换方法或信息发送方法。
附图说明
图1是本公开提供的DF选举模式的切换方法的流程示意图；
图2是本公开提供的DF选举模式的切换方法中部分步骤的流程示意图；
图3是本公开提供的DF选举模式的切换方法中部分步骤的流程示意图；
图4是本公开提供的DF选举模式的切换方法中部分步骤的流程示意图；
图5是本公开提供的DF选举模式的切换方法中部分步骤的流程示意图；
图6是本公开提供的DF选举模式的切换方法中部分步骤的流程示意图；
图7是本公开提供的信息发送方法的流程示意图；
图8是本公开提供的信息发送方法中部分步骤的流程示意图；
图9是本公开提供的信息发送方法中部分步骤的流程示意图；
图10是本公开提供的信息发送方法中部分步骤的流程示意图；
图11是本公开提供的信息发送方法中部分步骤的流程示意图；
图12是本公开提供的信息发送方法中部分步骤的流程示意图；
图13是本公开提供的DF选举模式切换应用场景的示意图；
图14是本公开提供的PE设备的示意图；以及
图15是本公开提供的计算机可存储介质的示意图。
具体实施方式
为使本领域的技术人员更好地理解本公开的技术方案，下面结合附图及具体实施例进行详细描述。
在下文中将参考附图更充分地描述示例实施例，但是所述示例实施例可以以不同形式来体现且不应当被解释为限于本文阐述的实施例。反之，提供这些实施例的目的在于使本公开透彻和完整，并将使本领域技术人员充分理解本公开的范围。
在不冲突的情况下，本公开各实施例及实施例中的各特征可相互组合。
如本文所使用的，术语“和/或”包括一个或多个相关列举条目的任何和所有组合。
本文所使用的术语仅用于描述特定实施例，且不意欲限制本公开。如本文所使用的，单数形式“一个”和“该”也意欲包括复数形式，除非上下文另外清楚指出。还将理解的是，当本说明书中使用术语“包括”和/或“由……制成”时，指定存在所述特征、整体、步骤、操作、元件和/或组件，但不排除存在或添加一个或多个其它特征、整体、步骤、操作、元件、组件和/或其群组。
除非另外限定，否则本文所用的所有术语(包括技术和科学术语)的含义与本领域普通技术人员通常理解的含义相同。还将理解，诸如那些在常用字典中限定的那些术语应当被解释为具有与其在相关技术以及本公开的背景下的含义一致的含义，且将不解释为具有理想化或过度形式上的含义，除非本文明确如此限定。
本公开实施例的基本网络场景为：同一用户边缘(CE，Customer Edge)设备以双归属方式接入第一PE设备与第二PE设备，在第一 PE设备设置有第一子接口，第二PE设备设置有与第一子接口相对应的第二子接口。当第二子接口的状态变化时，触发第一PE设备的第一子接口对应的DF选举模式的切换操作。
本公开提供了一种DF选举模式的切换方法，应用于第一PE设备，如图1所示，包括如下步骤S100至S200。
在步骤S100，响应于接收到第二PE设备发布的第一路由通告消息，将与第一子接口对应的DF选举模式切换到第一模式。
第一模式为没有第二PE设备为第一ES发布的以太网分段路由参与的DF选举模式。第一子接口及第二子接口均绑定了第一EVI，且第一子接口及第二子接口均为第一ES上的子接口。
第一路由通告消息是当第二子接口失效时由第二PE设备针对第二子接口的状态变化发布的携带多协议网络层可达信息MP_REACH_NLRI的路由通告消息。
第一PE设备收到第一路由通告消息后，会通过第一路由中包含的指示第一ES的ESI信息和指示第一EVI的EVI信息确定与之对应的第一子接口，从而将第二PE设备为第一ES发布的以太网分段路由排除在与第一子接口对应的DF选举流程之外。
在第二子接口失效而对应的主接口没有失效的情况下，采用第一模式，第一子接口对应的<ESI,EVI>上的DF选举会将第二子接口排除在外，从而避免该<ESI,EVI>的DF选举结果异常。将第二子接口排除在外，即，将第二子接口所在的PE节点(即，第二PE设备)发布的以太网分段路由排除在外。
需要说明的是，DF选举算法有多种，并且将来还可能出现新的DF选举算法，无论第一模式还是第二模式都可以采用这些算法，通过第一模式或第二模式和具体的DF选举算法(比如《draft-ietf-bess-evpn-pref-df》中的DF选举算法)的结合，才能得出最终的DF选举结果。
需要说明的是，如果是主接口失效，其所有子接口也会随之失效，此时，第二PE设备只需要撤销相应的以太网分段路由，并不需要对每个子接口发布第一路由通告消息。因此，第一PE设备根据收 到的第一路由通告消息，就可以知道第二PE设备上相应ES的相应子接口失效而其主接口并未失效。
第一PE设备接收到第二PE设备发布的第一路由通告消息后，将对应<ESI,EVI>上的DF选举模式由第二模式切换为第一模式。
在步骤S200，响应于接收到第二PE设备发布的第一路由撤销消息，将与第一子接口对应的DF选举模式切换到第二模式。
第二模式为有第二PE设备为第一ES发布的以太网分段路由参与的DF选举模式。
第一路由撤销消息是当第二子接口恢复正常时由第二PE设备针对第二子接口的状态变化发布的携带多协议网络层可达信息MP_UNREACH_NLRI的路由撤销消息。
第一PE设备收到第一路由撤销消息后，会通过第一路由中包含的指示第一ES的ESI信息和指示第一EVI的EVI信息确定与之对应的第一子接口，从而将第二PE设备为第一ES发布的以太网分段路由重新恢复参与到与第一子接口对应的DF选举流程中，就如同第二子接口失效之前那样，因此称第二模式为“有第二PE设备为第一ES发布的以太网分段路由参与的DF选举模式”。
在<ESI,EVI>范围内，可能存在多个网元，假如在多个网元上均发生了子接口故障，并且都向邻居发布了第一路由通告消息，则第一PE设备会从多个网元收到多个第一路由通告消息。因为每个第一路由撤销消息只能撤销其对应的一个第一路由通告消息，所以，不能在第一PE设备只收到一个第一路由撤销消息时就直接将与第一子接口对应的DF选举模式切换到第二模式，而是应在收到了与所有第一路由通告消息对应的各个撤销消息之后，才切换到第二模式。
第一PE设备接收到第二PE设备发布的第一路由撤销消息后，将对应<ESI,EVI>上的DF选举模式切换到步骤S100之前的第二模式。
每个<ESI,EVI>上的初始DF选举模式为第二模式(例如，非AC-DF模式)，不需要发布基于EVI的以太网自动发现(Ethernet Auto-Discovery per EVI，RT-1 per EVI)路由；当某个<ESI,EVI>上确实存在子接口故障时，该<ESI,EVI>上的DF选举模式才切换为第 一模式(例如，AC-DF模式)。此时，该<ESI,EVI>上才需要与发生故障的子接口量级相当的路由发布，每个<ESI,EVI>分别进入第一模式。当<ESI,EVI>上不存在故障时，按第一模式或第二模式进行DF选举都能得出正确结果，并且第二模式因其需要发布的路由数量成量级地减少，所以性能更佳。
如图2所示，在接收到第二PE设备发布的第一路由通告消息后，还可以包括步骤S101。
在步骤S101，响应于接收到第二PE设备发布的第一路由通告消息且第一子接口的状态正常，为第一子接口发布表示第一子接口的状态正常的路由通告消息。
需要说明的是，在步骤S101之前，第一PE设备并不需要出于DF选举的目的而向邻居发布专门针对具体某个没有故障的子接口的路由通告消息，从而能够成量级地减少路由通告消息。
由于在PBB EVPN等场景中，原本不发布基于EVI的以太网自动发现路由，在切换到第一模式前，并没有通过基于EVI的以太网自动发现路由获取到<ESI,EVI>范围内各个网元的子接口状态信息。因此，当收到第二PE设备发布的针对第二子接口失效的第一路由通告消息时，第一PE设备需要在第一路由<ESI,EVI>范围内，向EVPN邻居补发第一PE设备上的与第二子接口相对应的第一子接口正常工作的路由通告信息。该路由通告信息用于向邻居通告第一子接口的状态为正常。不限于第一PE设备，该<ESI,EVI>范围内所有网元都可以执行步骤S101，因此各个网元都能得知该<ESI,EVI>范围内所有正常工作的子接口所在的节点信息，从而将不能正常工作的子接口所在的节点排除在该<ESI,EVI>的DF选举流程之外。
需要说明的是，第一子接口与第二子接口属于同一<ESI,EVI>，且第一路由通告消息及第一路由撤销消息也是针对<ESI,EVI>。具体地，在示例中，都是针对<第一ESI，第一EVI>，因而它们之间可以通过<ESI,EVI>信息相互索引到。
在另一种实施方式中，出于节约网络资源的考虑，不需要执行步骤S101，而是由各网元均默认为已经知道除发来第一路由通告消 息的节点(即，第二PE设备)外，该<ESI,EVI>在其他所有节点上相应的子接口均是有效的，从而将相应子接口已经失效的节点排除在该<ESI,EVI>的DF选举流程之外。
需要说明的是，“将某节点排除在DF选举流程之外”是指将该节点发布的以太网分段路由排除在相应<ESI,EVI>的DF选举流程之外，即，排除其可以被选举为DF的可能性。
如图3所示，在接收到第二PE设备发布的第一路由通告消息之前，还包括步骤S102。
在步骤S102，响应于接收到第二PE设备发布的携带有第一指示信息的以太网分段路由，将第一ES中的每个<ESI,EVI>上的DF选举模式设为第二模式。
第一指示信息为在PE设备能够发布和接收第一路由的情况下发布的指示信息。
需要说明的是，通过第一指示信息，实施了本公开方案的节点可以确认网络中是否存在未实施本公开方案的节点。只有当同一ES上所有PE设备均实施了本公开方案时，实施了本公开方案的节点才真正对外发布第一路由通告消息。当第一PE设备收到第二PE设备发布的携带有第一指示信息的以太网分段路由时，会对是否收到第一路由通告消息进行判断，将第一ES中所有未收到第一路由通告消息的<ESI,EVI>上的DF选举设为第二模式，而将所有已收到第一路由通告消息的<ESI,EVI>上的DF选举设为第一模式。
如图4所示，接收到第二PE设备发布的第一路由通告消息后，还包括以下步骤S103。
在步骤S103，仅当收到的第一ES的所有以太网分段路由中均携带第一指示信息时，才将与第一子接口对应的DF选举模式切换到第一模式。
每个<ESI,EVI>上的初始DF选举模式为第二模式(例如，非AC-DF模式)，不需要发布基于EVI的以太网自动发现(Ethernet Auto-Discovery per EVI，RT-1 per EVI)路由；当某个<ESI,EVI>上确实存在子接口故障时，该<ESI,EVI>上的DF选举模式才切换为第 一模式(例如，AC-DF模式)。此时，该<ESI,EVI>上才需要与基于EVI的以太网自动发现路由量级相当的路由(比如，第一路由)发布，每个<ESI,EVI>分别进入第一模式。当<ESI,EVI>上不存在故障时，按第一模式或第二模式进行DF选举都能得出正确结果，并且第二模式因其需要发布的路由数量成量级地减少，所以性能更佳。
如图5所示，接收到第二PE设备发布的第一路由撤销消息后，还包括以下步骤S104。
在步骤S104，仅当收到的第一ES的所有以太网分段路由中均携带第一指示信息时，才将与第一子接口对应的DF选举模式切换到第二模式。
与步骤S103相对应，在第二子接口恢复正常时，需要通过第一路由撤销消息通知相应<ESI,EVI>上的DF选举模式切换到步骤S103之前的第二模式。
响应于接收到第二PE设备发布的第一路由通告消息，将第二PE设备为第一ES发布的以太网分段路由排除在与第一子接口对应的DF选举流程之外。第一路由可以为额外携带第二指示信息的基于EVI的以太网自动发现路由，其中，第二指示信息可以在EVPN二层属性扩展团体(Layer 2 Attributes Extended Community)属性中携带，具体地，可以在该属性的“控制标记(Control Flags)”字段中的某个标志位中携带。为描述方便起见，将携带第二指示信息的标志位称为AC失效标志位。当第二接口失效时，由第二PE设备发布这种格式的以太网自动发现路由通告消息，其中AC失效标志位的值设置为“1”，用以告诉接收端PE设备其为第一路由通告消息。
可选地，AC失效标志位可以不限于1位，也不限于设置为“1”时表示AC失效，只要是包含AC失效含义的设置，都在本公开保护范围之内。
当第一PE设备收到第一路由通告消息后，将相应的以太网分段路由排除在DF选举流程之外。
在接收第一路由通告消息之前，可以通过边界网关协议(BGP，Border Gateway Protocol)的路由目标约束(Route Target  Constraints，RT Constraints)机制向邻居通告包含第一路由目标的路由目标成员资格(Route Target membership)网络层可达性信息(NLRI，Network Layer Reachability Information)。
通常情况下，第二PE设备的EVPN路由在发布时不被限制，会携带路由目标发往所有邻居，接收侧根据路由目标进行匹配，只接受可以匹配的路由。通过路由目标对ES范围进行限定，可以避免在不必要的邻居上占用路由资源。
路由目标可采用特殊类型，比如ES入方向路由目标(ES Import RT)和EVI路由目标(EVI-RT)，目的是在第一路由通告消息中同时携带了第一ES的路由目标和第一EVI的路由目标时，接收第一路由通告消息的节点可以将两种路由目标区分开，从而将第一路由的导入范围控制在同一ES内的PE设备上，而网络中其它无关的PE设备则不会接受带这些路由目标的路由，并且只有ES Import RT和EVI-RT均匹配的情况下才能导入。
在路由转发路径上，可能存在运行RT Constraints机制的节点(比如，某个路由反射器(RR，Route Reflector)节点)，将会对第二PE设备发布的EVPN路由做限制。为避免RT Constraints机制带来的影响，在接收第一路由通告消息之前，通过BGP的RT Constraints机制向邻居通告包含第一路由目标的路由目标成员资格NLRI，可以使携带有第一路由目标的第一路由在网络中顺利转发。
如图6所示，还可以包括以下步骤S105。
在步骤S105，响应于接收到第二PE设备发布的带有限定第一ES的路由目标的第一路由通告消息，且第三PE设备本地不存在与路由目标匹配的ES，第三PE设备忽略该第一路由通告消息。
第二PE设备与第三PE设备存在EVPN邻居关系，且第三PE设备的子接口均不在第一ES上。当第二PE设备发布的第一路由通告消息中带有限定第一ES的路由目标时，因为第三PE设备的子接口均不在第一ES上，所以第三PE设备本地不存在与路由目标匹配的ES，因此忽略该第一路由通告消息。
本公开还提供了一种信息发送方法，应用于第二PE设备，如图 7所示，包括如下步骤S300至S400。
在步骤S300，响应于第二子接口失效，发布第一路由通告消息。
第一子接口及第二子接口均绑定了第一EVI，且第一子接口及第二子接口均为第一ES上的子接口。
当ESI的主接口没有失效而其子接口失效时，需要通知邻居网元，在对应该子接口<ESI,EVI>上的DF选举流程中将失效子接口(例如，第二子接口)所在的节点(即，发布第一路由通告消息的节点)排除在外，否则<ESI,EVI>上的DF选举结果会异常。
第一路由通告消息是当第二子接口失效时由第二PE设备针对第二子接口的状态变化发布的携带多协议网络层可达信息MP_REACH_NLRI的路由通告消息。第一路由可以采用在已有路由类型中添加特殊标志的方式，也可以采用新增的地址族的方式，或采用新增路由类型的方式。比如，采用路由类型编码大于13的新的EVPN路由类型。
虽然第一路由通告消息针对的是子接口“失效”，但本公开采用的是多协议网络层可达信息MP_REACH_NLRI，而不是多协议网络层不可达信息MP_UNREACH_NLRI。原因在于，MP_REACH_NLRI是针对路由通告，而MP_UNREACH_NLRI是针对MP_REACH_NLRI的撤销，如果之前本就没有收到过与MP_UNREACH_NLRI对应的MP_REACH_NLRI，则携带MP_UNREACH_NLRI的路由撤销消息是无效的消息。通常做法是将携带该MP_UNREACH_NLRI的路由撤销消息丢弃。
比如，当第二PE设备与第一PE设备之间需要由RR转发路由消息时，假设第二PE设备用携带MP_UNREACH_NLRI的路由撤销消息来通告第二子接口失效，RR本就没收到过与该MP_UNREACH_NLRI对应的MP_REACH_NLRI，所以当RR收到携带该MP_UNREACH_NLRI的路由撤销消息时，会丢弃此消息，而不会将此消息转发给第一PE设备。
本公开中的第一路由是将子接口的“失效”作为路由消息的内容通告给邻居，子接口的“失效”与MP_REACH_NLRI的“可达”不是一个概念。仍以上述RR场景为例，第二子接口失效时，第二PE设备用携带MP_REACH_NLRI的路由通告消息来通告第二子接口失效，RR 收到该路由通告消息后，会将该路由通告消息转发给第一PE设备。
在步骤S400，响应于第二子接口恢复正常，发布第一路由撤销消息。
第二子接口恢复正常时，第一路由使用携带MP_UNREACH_NLRI的路由撤销消息来撤销在步骤S300中子接口“失效”时发布的相应携带MP_REACH_NLRI的路由通告消息。仍以步骤S300中的RR场景为例，第二子接口恢复正常时，第二PE发布携带MP_UNREACH_NLRI的路由撤销消息，用来撤销此前发布的与之对应的携带MP_REACH_NLRI的路由通告消息，RR收到携带MP_UNREACH_NLRI的路由撤销消息后，发现此前曾收到过与之对应的携带MP_REACH_NLRI的路由通告消息，因此不会直接丢弃，会将此消息转发给第一PE设备。
如图8所示，还可以包括以下步骤S301。
在步骤S301，在发布第一路由通告消息之前，发布携带有第一指示信息的以太网分段路由。
第一指示信息为在PE设备能够发布和接收第一路由的情况下发布的指示信息。
可以扩展DF选举扩展团体(Extended Community)属性，在其中的“Bitmap”字段中扩展一个表示基于EVI切换到AC-DF模式的标志位，作为第一指示信息的一种实现方式，以表示第二PE设备实施了本公开方案(即，能够发送和接收第一路由)。此时称第二PE设备对第一ES执行基于EVI的AC-DF(称为AC-DF per EVI)模式。所谓基于EVI的AC-DF模式就是：起始态(即，没有任何故障的状态)该ES上每个<ESI,EVI>均采用第二模式，而当感知到某个<ESI,EVI>故障之后，再单独在该<ESI,EVI>粒度上使能第一模式，当该<ESI,EVI>故障解除之后，恢复第二模式。
需要说明是，基于EVI的AC-DF模式指的是某个ES上各个<ESI,EVI>采用的DF选举模式(即，第一模式或第二模式)的整体情况，而第一模式和第二模式指的是该ES的某个具体<ESI,EVI>上的DF选举模式。从这个意义上讲，第一指示信息还表示了相应ESI的各个<ESI,EVI>采用的DF选举模式的整体情况为基于EVI的AC-DF 模式。第一指示信息表示第二PE设备实施了本公开方案，或者第一指示信息表示第二PE设备能够发送和接收第一路由，或者第一指示信息表示第二PE设备所在的第一ES的各个<ESI,EVI>采用的DF选举模式的整体情况为基于EVI的AC-DF模式，是同一种意思的不同表达方式。
需要说明的是，当所有以太网分段路由均携带第一指示信息时，说明相应ES上各PE节点均实施了本公开方案，此时，可以说在ES级别上协商为支持第一路由，或者也可以说在ES粒度上将DF选举模式协商为基于EVI的AC-DF模式，或者也可以说该ES的各个<ESI,EVI>采用的DF选举模式的整体情况协商为基于EVI的AC-DF模式，这几种说法都是同一意思的不同表达方式。
在一些实施例中，第一指示信息可以与AC-DF能力(AC-DF Capability)标志同时设置。
在扩展团体属性的Bitmap字段中，DF选举模式标志中第一指示信息与AC-DF能力标志位可以同时置位。
当第一指示信息与该Bitmap字段中AC-DF能力标志同时置位时，如果ES的所有以太网分段路由均携带了第一指示信息(说明该ES中不存在未实施本公开方案的节点)，则AC-DF能力标志自动失效；否则AC-DF能力标志仍然有效。这样可以确保在ES中存在不支持第一指示信息的节点时，仍然可以按AC-DF能力标志指示的方法选举DF，从而能更好地发挥支持AC-DF但未实施本公开方案的节点的DF选举能力。
如图9所示，响应于第二子接口失效，可以包括以下步骤S302。
在步骤S302，仅当所有PE设备发布的以太网分段路由中均携带第一指示信息时，才分布第一路由通告消息。
第一指示信息为在PE设备能够发布和接收第一路由的情况下发布的指示信息。
第一路由消息可以不新增路由类型，仍采用已有的基于EVI的以太网自动发现路由消息，向第一PE设备通告第二子接口的状态变化。这样做的好处是，更容易与现有网络(比如其中的RR节点)兼 容。
需要说明的是，通过第一指示信息，实施了本公开方案的节点可以确认网络中是否存在未实施本公开方案的节点。只有当同一ES上所有PE设备均实施了本公开方案时，实施了本公开方案的节点才真正对外发布第一路由通告消息。这样，当网络中存在未实施本公开方案(即，未在以太网分段路由中携带第一指示信息)的节点时，可以按照相应ES的所有以太网分段路由均未携带第一指示信息时的规则进行处理，从而保证了在新旧节点混合组网时，不会使结果变得更糟。
如图10所示，响应于第二子接口失效，还可以包括以下步骤S303。
在步骤S303，在第一路由通告消息中，将多协议标签交换协议标签(MPLS Label，Multiprotocol Label Switching Protocol Label)字段设为保留值，将第一路由通告消息的以太网标签标识(Ethernet Tag ID)字段设为与第一EVI对应的I组件服务标识(I-SID，I-Component Service ID)值，且在第一路由通告消息中携带第一骨干组件(B-Component，Backbone Component)的路由目标，其中，第一骨干组件为与第一EVI绑定的骨干组件。
在步骤S301的基础上，为了区别于MP_REACH_NLRI中的常规设定，在基于EVI的以太网自动发现路由中，对MP_REACH_NLRI做特殊设置，包括将MPLS Label字段设为不与现有协议冲突的保留值，比如0，且将Ethernet Tag ID字段设为与第一EVI对应的I-SID值。
在一些实施例中，在第一路由通告消息中携带用于标识该路由为第一路由的第二指示信息。
第一路由可以是新增的路由类型(此时，路由类型的取值即是所述第二指示信息)，也可以是在基于EVI的以太网自动发现路由上进行了特殊设置的路由。
第一路由可以使用L2VPN/EVPN地址族，其NLRI类型为1，在EVPN二层属性扩展团体属性的Control Flags字段中添加AC失效标志作为第二指示信息。接收侧可以根据扩展团体属性中的AC失效标 志得知第二PE设备的第二子接口失效，并将第二PE设备为第一ES发布的以太网分段路由排除在与第一子接口对应的DF选举流程之外。
如图11所示，响应于第二子接口失效，还可以包括以下步骤S304。
在步骤S304，在第一路由通告消息中携带与第一ES对应的第一路由目标。
通常情况下，第二PE设备的EVPN路由在发布时不被限制，会携带路由目标发往所有邻居，接收侧根据路由目标进行匹配，只接受可以匹配的路由。通过路由目标对ES范围进行限定，可以避免在不必要的邻居上占用路由资源。
第一路由可以是新增的路由类型，也可以是在基于EVI的以太网自动发现路由上进行了特殊设置的路由。MP_REACH_NLRI字段的填写方法与基于EVI的以太网自动发现路由相同，但其路由目标采用特殊类型，比如ES Import RT和EVI-RT，目的是在第一路由通告消息中同时携带了第一ES的路由目标和第一EVI的路由目标时，接收第一路由通告消息的节点可以将两种路由目标区分开，从而将第一路由的导入范围控制在同一ES内的PE设备上，而网络中其它无关的PE设备则不会接受带这些路由目标的路由。此时，ES Import RT和/或EVI-RT的出现即为所述第二指示信息。
第一路由目标可以对RT Constraints机制可见。
在路由转发路径上，可能运行有RT Constraints机制，将会对第二PE设备发布的EVPN路由做限制。为避免RT Constraints机制带来的影响，在接收第一路由通告消息之前，通过BGP的RT Constraints机制向邻居通告包含第一路由目标的路由目标成员资格NLRI。第一路由目标对RT Constraints机制可见，即，RT Constraints机制会检查第一路由目标，若第一路由目标不能匹配路由目标成员资格，则带有第一路由目标的路由将不被发布。
第一路由目标可以用于使得不在第一ES上的PE节点对第一路由通告消息进行过滤。
第一路由通告消息同时携带第一ES的路由目标和第一EVI的路 由目标。由于PE节点(例如，第三PE设备)不在第一ES上，因此第一ES的路由目标无法匹配，并忽略第一路由通告消息。
如图12所示，响应于第二子接口失效，还可以包括以下步骤S305。
在步骤S305，在第一路由通告消息中携带与第一EVI对应的第二路由目标，且第二路由目标对RT Constraints机制不可见。
通过将第一EVI对应的第二路由目标设定为对RT Constraints机制不可见，即使第三PE设备通过RT Constraints机制表示其希望接收携带该路由目标的路由，第一路由通告消息也会由于与第一ES对应的路由目标不匹配，不会被转发给第三PE设备，从而保证了在部署了RT Constraints的网络中，第一路由通告消息只会到达第一ES上的PE节点。比如，第三PE设备由于配置有第一EVI，其会发布针对第一EVI的路由目标(即，第二路由目标)的路由目标成员资格NLRI，但不会发布针对第一ES的路由目标(即，第一路由目标)的路由目标成员资格NLRI，因此，虽然第一路由通告消息中包含了对RT-Constraints机制不可见的第二路由目标，第一路由通告消息也会因为第一路由目标没有匹配到第三PE设备发布的路由目标成员资格NLRI而不对第三PE发布。
本公开提供了一个具体的实例，如图13所示，PE1、PE2分别为第一PE设备、第二PE设备(不限定必须是PE1为第一PE设备、PE2为第二PE设备，具体角色对应关系根据实际情况而定)，PE3为示例性的其他PE设备(比如第三PE设备)，R4和R5分别为示例性的PE设备之间的其他设备，CE设备所在的ES的ESI为ESI2，ESI2绑定在主接口IF_1上，但业务绑定在IF_1的子接口IF_1.1和IF_1.2上。子接口IF_1.1和IF_1.2作为AC绑定在I组件(即，EVPN实例，EVI)中，第一路由在I组件中发布，且不发布以太网自动发现路由。
I组件是PBB EVPN的I组件，除I组件外，PBB EVPN还有B组件(即，骨干组件(B-Component))，示例中示出了两个EVPN实例(EVI)，分别是100和200，每个EVI在每台PE设备上各有一个对应的I组件，DF选举要为每个<ESI,EVI>选出唯一的DF结点。比如， 对于<ESI2,EVI100>，DF选举结果为：PE1为DF、PE2为非DF；对于<ESI2,EVI200>，DF选举结果为：PE1为非DF、PE2为DF。
缺省情况下，ESI2在各PE设备上的主接口IF_1及其所有子接口均是UP的，因此，对ESI2的DF选举采用第二模式，也就是说，当前DF选举结果不受其中不包含第二指示信息的基于EVI的以太网自动发现路由是否存在的影响。
当PE1上子接口IF_1.1失效而其主接口以及其它子接口没有失效时，PE1针对子接口IF_1.1发布第一路由通告消息，同一ES上的其它PE设备(以PEn为例)可以通过该第一路由通告消息将相应<ESI,EVI>上的DF选举从第二模式切换到第一模式。
当PE1上子接口IF_1.1恢复后，PE1发布该第一路由的撤销消息，当PEn收到该撤销消息时，消除此前从PE1收到的第一路由通告消息，并且将该<ESI,EVI>上的DF选举从第一模式切换到第二模式。如果同一<ESI,EVI>上不再存在其它第一路由通告消息，则所有PE设备为该ESI发布的以太网分段路由都将参与针对该<ESI,EVI>的DF选举流程。
除PBB EVPN外，本公开还可应用于虚拟可扩展局域网(VXLAN，Virtual Extensible Local Area Network)EVPN，或分段路由IPv6(SRv6,Segment Route IPv6)EVPN的不发布以太网自动发现路由的场景。在这些场景中，可以不采用MAC/IP地址发布路由(MAC/IP Advertisement Route)，而是在数据面学习远端MAC条目来取代MAC/IP地址发布路由的作用。
需要说明的是，在PBB EVPN场景中，第一路由通告消息可以只携带B组件的路由目标而不携带I组件的路由目标，此时，可以通过将第一路由通告消息中的Ethernet Tag ID字段设置为相应I-SID来区分针对不同I组件的第一路由通告消息。
值得注意的是，第一路由可以是一种新的EVPN路由类型(当第二指示信息为其路由类型时)，其与以太网自动发现路由的发布、撤销时机是相反的，即，后者的发布时机是第一路由的撤销时机，后者的撤销时机是第一路由的发布时机，因此，第一路由是在故障时发布， 故障恢复时撤销的，由于故障是个别的，故而常态下不会发布第一路由，也就是说，常态下第一路由不会占用网络资源。只有当某个<ESI,EVI>上确实存在子接口故障时，DF选举才会需要切换为第一模式，才需要发布与发生故障的子接口量级相当的第一路由，而不需要发布与未发生故障的子接口量级相当的路由。由于发生故障的子接口数量远远小于未发生故障的子接口数量，在第一模式需要发布的路由数量成量级地减少，在第二模式更是不需要发布与子接口量级相当的路由，降低了系统开销，节约了网络资源。
如图14所示，本公开提供了一种PE设备，包括：一个或多个处理器501；以及存储器502，其上存储有一个或多个程序，当一个或多个程序被一个或多个处理器501执行，使得一个或多个处理器501实现根据本公开各实施例的DF选举模式的切换方法或信息发送方法。
此外，还可以包括一个或多个I/O接口503，其连接在处理器501与存储器502之间，并且配置为实现处理器501与存储器502的信息交互。
处理器501可以为具有数据处理能力的器件，其包括但不限于中央处理器(CPU)等；存储器502可以为具有数据存储能力的器件，其包括但不限于随机存取存储器(RAM，更具体如SDRAM、DDR等)、只读存储器(ROM)、带电可擦可编程只读存储器(EEPROM)、闪存(FLASH)；I/O接口503连接在处理器501与存储器502间，能实现处理器501与存储器502的信息交互，其包括但不限于数据总线(Bus)等。
处理器501、存储器502和I/O接口503可以通过总线504相互连接，进而与计算设备的其它组件连接。
如图15所示，本公开提供一种计算机可读介质，其上存储有计算机程序，程序被处理器执行时实现根据本公开各实施例的DF选举模式的切换方法或信息发送方法。
本领域普通技术人员可以理解，上文中所公开方法中的全部或某些步骤、系统、装置中的功能模块/单元可以被实施为软件、固件、 硬件及其适当的组合。在硬件实施方式中，在以上描述中提及的功能模块/单元之间的划分不一定对应于物理组件的划分；例如，一个物理组件可以具有多个功能，或者一个功能或步骤可以由若干物理组件合作执行。某些物理组件或所有物理组件可以被实施为由处理器，如中央处理器、数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术语计算机存储介质包括在用于存储信息(诸如计算机可读指令、数据结构、程序模块或其它数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其它存储器技术、CD-ROM、数字多功能盘(DVD)或其它光盘存储、磁盒、磁带、磁盘存储或其它磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其它的介质。此外，本领域普通技术人员公知的是，通信介质通常包含计算机可读指令、数据结构、程序模块或者诸如载波或其它传输机制之类的调制数据信号中的其它数据，并且可包括任何信息递送介质。
本文已经公开了示例实施例，并且虽然采用了具体术语，但它们仅用于并仅应当被解释为一般说明性含义，并且不用于限制的目的。在一些实例中，对本领域技术人员显而易见的是，除非另外明确指出，否则可单独使用与特定实施例相结合描述的特征、特性和/或元素，或可与其它实施例相结合描述的特征、特性和/或元件组合使用。因此，本领域技术人员将理解，在不脱离由所附的权利要求阐明的本公开的范围的情况下，可进行各种形式和细节上的改变。

Claims (16)

  1. 一种指定转发者DF选举模式的切换方法，应用于第一提供商边缘PE设备，所述第一PE设备与第二PE设备均在第一以太网分段ES上，所述第一PE设备设置有第一子接口，所述第二PE设备设置有与所述第一子接口相对应的第二子接口，所述第一子接口及所述第二子接口均绑定了第一以太网虚拟专用网络实例EVI，且所述第一子接口及所述第二子接口均为所述第一ES上的子接口，所述方法包括：
    响应于接收到所述第二PE设备发布的第一路由通告消息，将与所述第一子接口对应的DF选举模式切换到第一模式，其中，所述第一模式为没有所述第二PE设备为所述第一ES发布的以太网分段路由参与的DF选举模式；以及
    响应于接收到所述第二PE设备发布的第一路由撤销消息，将与所述第一子接口对应的DF选举模式切换到第二模式，其中，所述第二模式为有所述第二PE设备为所述第一ES发布的以太网分段路由参与的DF选举模式，
    其中，所述第一路由为所述第二PE设备针对所述第二子接口的状态变化而发布的路由，所述第一路由通告消息是当所述第二子接口失效时由所述第二PE设备针对所述第二子接口的状态变化发布的携带多协议网络层可达信息MP_REACH_NLRI的路由通告消息，并且所述第一路由撤销消息是当所述第二子接口恢复正常时由所述第二PE设备针对所述第二子接口的状态变化发布的携带多协议网络层不可达信息MP_UNREACH_NLRI的路由撤销消息。
  2. 根据权利要求1所述的方法，还包括：
    响应于接收到所述第二PE设备发布的所述第一路由通告消息且所述第一子接口的状态正常，为所述第一子接口发布表示所述第一子接口的状态正常的路由通告消息。
  3. 根据权利要求1所述的方法，其中，在接收到所述第二PE设 备发布的所述第一路由通告消息之前，还包括：
    响应于接收到所述第二PE设备发布的携带有第一指示信息的以太网分段路由，将所述第一ES中的每个<ESI,EVI>上的DF选举模式设为所述第二模式，
    其中，所述第一指示信息为在PE设备能够发布和接收所述第一路由的情况下发布的指示信息。
  4. 根据权利要求1所述的方法，其中，
    响应于接收到所述第二PE设备发布的第一路由通告消息，将与所述第一子接口对应的DF选举模式切换到所述第一模式包括：
    仅当收到的所述第一ES的所有以太网分段路由中均携带第一指示信息时，才将与所述第一子接口对应的DF选举模式切换到所述第一模式；
    响应于接收到所述第二PE设备发布的第一路由撤销消息，将与所述第一子接口对应的DF选举模式切换到所述第二模式包括：
    仅当收到的所述第一ES的所有以太网分段路由中均携带所述第一指示信息时，才将与所述第一子接口对应的DF选举模式切换到所述第二模式；
    其中，所述第一指示信息为在PE设备能够发布和接收所述第一路由的情况下发布的指示信息。
  5. 根据权利要求1至4中任意一项所述的方法，其中，所述第二PE设备与第三PE设备存在以太网虚拟专用网络EVPN邻居关系，且所述第三PE设备的子接口均不在所述第一ES上，所述方法还包括：
    响应于接收到所述第二PE设备发布的带有限定所述第一ES的路由目标的所述第一路由通告消息，且所述第三PE设备本地不存在与所述路由目标匹配的ES，所述第三PE设备忽略所述第一路由通告消息。
  6. 一种信息发送方法，应用于第二提供商边缘PE设备，所述第 二PE设备与第一PE设备均在第一以太网分段ES上，所述第一PE设备设置有第一子接口，所述第二PE设备设置有与所述第一子接口相对应的第二子接口，所述第一子接口及所述第二子接口均绑定了第一以太网虚拟专用网络实例EVI，且所述第一子接口及所述第二子接口均为所述第一ES上的子接口，所述方法包括：
    响应于所述第二子接口失效，发布第一路由通告消息；以及
    响应于所述第二子接口恢复正常，发布第一路由撤销消息，
    其中，所述第一路由为所述第二PE设备针对所述第二子接口的状态变化而发布的路由，所述第一路由通告消息是当所述第二子接口失效时由所述第二PE设备针对所述第二子接口的状态变化发布的携带多协议网络层可达信息MP_REACH_NLRI的路由通告消息，并且所述第一路由撤销消息是当所述第二子接口恢复正常时由所述第二PE设备针对所述第二子接口的状态变化发布的携带多协议网络层不可达信息MP_UNREACH_NLRI的路由撤销消息。
  7. 根据权利要求6所述的方法，还包括：
    在发布所述第一路由的通告消息之前，发布携带有第一指示信息的以太网分段路由，
    其中，所述第一指示信息为在PE设备能够发布和接收所述第一路由的情况下发布的指示信息。
  8. 根据权利要求6所述的方法，其中，响应于所述第二子接口失效，发布所述第一路由通告消息包括：
    仅当所有PE设备发布的以太网分段路由中均携带第一指示信息时，才发布所述第一路由通告消息，
    其中，所述第一指示信息为在PE设备能够发布和接收所述第一路由的情况下发布的指示信息。
  9. 根据权利要求8所述的方法，还包括：
    在所述第一路由通告消息中，将多协议标签交换协议标签MPLS  Label字段设为保留值，将所述第一路由通告消息的以太网标签标识Ethernet Tag ID字段设为与所述第一EVI对应的I组件服务标识I-SID值，且在所述第一路由通告消息中携带第一骨干组件的路由目标，其中，所述第一骨干组件为与所述第一EVI绑定的骨干组件。
  10. 根据权利要求6至9中任意一项所述的方法，还包括：
    在所述第一路由通告消息中携带与所述第一ES对应的第一路由目标。
  11. 根据权利要求10所述的方法，其中：
    所述第一路由目标对路由目标约束机制可见；或者
    所述第一路由目标用于使得不在第一ES上的PE设备对所述第一路由通告消息进行过滤。
  12. 根据权利要求6至9中任意一项所述的方法，还包括：
    在所述第一路由通告消息中携带与所述第一EVI对应的第二路由目标，且所述第二路由目标对路由目标约束机制不可见。
  13. 一种提供商边缘PE设备，包括：
    一个或多个处理器；以及
    存储器，其上存储有一个或多个程序，当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现根据权利要求1至5中任意一项所述的DF选举模式的切换方法。
  14. 一种提供商边缘PE设备，包括：
    一个或多个处理器；以及
    存储器，其上存储有一个或多个程序，当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现根据权利要求6至12中任意一项所述的信息发送方法。
  15. 一种计算机可读介质，其上存储有计算机程序，所述程序被处理器执行时实现根据权利要求1至5中任意一项所述的DF选举模式的切换方法。
  16. 一种计算机可读介质，其上存储有计算机程序，所述程序被处理器执行时实现根据权利要求6至12中任意一项所述的信息发送方法。
PCT/CN2022/082171 2021-04-30 2022-03-22 Df选举模式切换方法、信息发送方法、pe设备和介质 WO2022227940A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110481920.X 2021-04-30
CN202110481920.XA CN115277302A (zh) 2021-04-30 2021-04-30 Df选举模式切换方法、信息发送方法、pe设备、介质

Publications (1)

Publication Number Publication Date
WO2022227940A1 true WO2022227940A1 (zh) 2022-11-03

Family

ID=83746064

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/082171 WO2022227940A1 (zh) 2021-04-30 2022-03-22 Df选举模式切换方法、信息发送方法、pe设备和介质

Country Status (2)

Country Link
CN (1) CN115277302A (zh)
WO (1) WO2022227940A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106487675A (zh) * 2015-08-31 2017-03-08 丛林网络公司 用于evpn中具有链路故障时的bum流量的出口保护
US20170195210A1 (en) * 2015-12-30 2017-07-06 Juniper Networks, Inc. Static route advertisement
WO2017118880A1 (en) * 2016-01-08 2017-07-13 Telefonaktiebolaget Lm Ericsson (Publ) Faster convergence on primary provider edge (pe) failure in a single-active redundancy topology
CN107666397A (zh) * 2016-07-29 2018-02-06 丛林网络公司 Pe路由器之间传送多播组离开请求的方法和pe路由器
CN108600070A (zh) * 2018-03-30 2018-09-28 新华三技术有限公司 指定转发者df选举方法及装置
CN112688873A (zh) * 2019-10-18 2021-04-20 瞻博网络公司 在evpn中部署安全邻居发现

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106487675A (zh) * 2015-08-31 2017-03-08 丛林网络公司 用于evpn中具有链路故障时的bum流量的出口保护
US20170195210A1 (en) * 2015-12-30 2017-07-06 Juniper Networks, Inc. Static route advertisement
WO2017118880A1 (en) * 2016-01-08 2017-07-13 Telefonaktiebolaget Lm Ericsson (Publ) Faster convergence on primary provider edge (pe) failure in a single-active redundancy topology
CN107666397A (zh) * 2016-07-29 2018-02-06 丛林网络公司 Pe路由器之间传送多播组离开请求的方法和pe路由器
CN108600070A (zh) * 2018-03-30 2018-09-28 新华三技术有限公司 指定转发者df选举方法及装置
CN112688873A (zh) * 2019-10-18 2021-04-20 瞻博网络公司 在evpn中部署安全邻居发现

Also Published As

Publication number Publication date
CN115277302A (zh) 2022-11-01

Similar Documents

Publication Publication Date Title
EP3836490B1 (en) Vpn cross-domain implementation method, device, and border node
US9019814B1 (en) Fast failover in multi-homed ethernet virtual private networks
US9992154B2 (en) Layer 3 convergence for EVPN link failure
CN108574616B (zh) 一种处理路由的方法、设备及系统
US9781032B1 (en) MPLS label usage in ethernet virtual private networks
US9858163B2 (en) Dual adjacency between edge devices at a network site
US8724456B1 (en) Network path selection for multi-homed edges to ensure end-to-end resiliency
WO2021258754A1 (zh) 报文指示方法、装置、设备和存储介质
JP7176095B2 (ja) 通信方法、通信機器、および通信システム
JP2010509825A (ja) リンクステートプロトコル制御のＥｔｈｅｒｎｅｔネットワークでのトラヒックエンジニアリングされたパス
CA3104756C (en) Loop avoidance communications method, device, and system
CN113132235B (zh) 基于虚电路的数据报文处理方法、转发表项的构建方法
CN113615132A (zh) 快速泛洪拓扑保护
WO2020244304A1 (zh) 路由信息发送的方法、路由选路的方法和装置
WO2022227940A1 (zh) Df选举模式切换方法、信息发送方法、pe设备和介质
CN114531396B (zh) 一种以太网虚拟专用网络中的故障回切方法及装置
US10735252B2 (en) Outside router fault detection
WO2022127284A1 (zh) 信息处理方法、节点及计算机可读存储介质
EP4016965A1 (en) Information processing method, network controller, node and computer-readable storage medium
US11212221B1 (en) Methods to improve ECMP for BGP services and BGP for provider edges with hybrid label distribution
WO2022246693A1 (en) Method and apparatus for path switchover management
WO2024001633A1 (zh) 网络管理方法及装置、网元、计算机可读存储介质
CN115037677A (zh) 一种frr场景保护的方法和装置
CN115460107A (zh) 路由检测方法、设备、系统及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22794416

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE