Abstract: A label management method, a data stream processing method, and a device, where the label management method includes allocating, by a controller, a source label to a data stream, sending, by the controller, a first Border Gateway Protocol (BGP) update packet to an ingress network device on a label switching path (LSP) of the data stream, and sending a second BGP update packet to an egress network device on the LSP of the data stream, where the first BGP update packet includes a stream identifier of the data stream and the source label, and the second BGP update packet includes a mapping relationship between the source label and a source object of the data stream. Hence, configuration, management, and maintenance can be simplified.

Abstract: The present disclosure discloses a method and an apparatus for implementing load sharing. The method includes: for a congested first link on a first forwarding node, selecting, by a network device, a packet flow forwarded by using the first link; selecting a second link that may be used to forward the packet flow and that is not congested after available bandwidth of the second link is occupied by the packet flow, where the second link is a link between the first forwarding node and a second forwarding node; selecting a first hash gene corresponding to the second link; determining that a third link is not in a congested state after available bandwidth of the third link is occupied by the packet flow; and saving the first hash gene in a source node of the packet flow, where the third link is a link that is on the second forwarding node.

Abstract: The present disclosure discloses a packet processing method, device, and system. The system includes: a controller, configured to: allocate a service label to a service processing manner of an FEC, establish a mapping relationship between the service label and the service processing manner, send the service label to a source node, and send the mapping relationship to a destination node; the source node, configured to: receive the service label sent by the controller, receive a first packet, insert the service label to the first packet to obtain a second packet, and send the second packet to the destination node; the destination node, configured to: receive the mapping relationship sent by the controller, receive the second packet sent by the source node, and pop the service label from the second packet according to the mapping relationship, to obtain the first packet.

Abstract: A method for allocating a global label is provided, including: receiving, by a controller, a first message sent by a forwarding apparatus, where the first message carries a first label range, multiple labels included in the first label range are a subset of multiple labels included in a label space of the forwarding apparatus, and the first label range is not equal to the label space of the forwarding apparatus; and allocating, by the controller, a global label to the forwarding apparatus according to the first label range carried in the first message, where the global label is a label in the first label range. In addition, a method for obtaining a global label, a controller, and a forwarding apparatus are further provided. The foregoing solutions improve efficiency of global label allocation.

Abstract: A network system includes one controller and a plurality of forwarding devices, where the forwarding devices forms a plurality of network topologies; the controller is configured to: allocate, according to a different attribute requirement of a virtual link on a basic link in the network topology, a link global label to the basic link based on the attribute requirement, and send a link binding relationship between the link global label and <a network topology ID, a basic link identifier, and an attribute> to the forwarding devices, where global labels allocated to each virtual link are different; and the forwarding devices is configured to receive the link binding relationship, and generate a label forwarding entry including a mapping relationship between the link global label and virtual link forwarding information, where the virtual link forwarding information includes the basic link and specified attribute processing that are corresponding to the global label.

Abstract: A network system includes one controller and a plurality of forwarding devices, where the forwarding devices form a plurality of virtual network topologies; the controller is configured to allocate a global label to each of the plurality of virtual network topologies, and send a binding relationship between the global labels and the virtual network topologies to the forwarding devices, where global labels allocated to different virtual network topologies are different; the forwarding devices are configured to receiving the binding relationship, and generate label forwarding entries including a mapping relationship between the global labels and forwarding information bases; and when a packet received by the forwarding device carries a global label, the forwarding device searches according to the global label, a forwarding information base of a virtual network topology corresponding to the global label, and determines that the packet is to be forwarded by using the forwarding information base.

Abstract: A label distribution method and device are disclosed. In an embodiment the method includes: a first routing device sends a first BGP update message including a first VPN instance identifier and a first RT to a second routing device; the first routing device receives a second BGP update message including a second VPN instance identifier and a second RT; the first routing device allocates a first VPN instance label to a second VPN instance after the second routing device determines that the first RT is equal to the second RT; and the first routing device sends a third BGP update message including the first VPN instance label to the second routing device. Mutual learning on VPN instances is performed, and a label is allocated to a VPN instance, so that a service tail end node can identify a VPN instance from which an L3VPN service specifically comes.

Abstract: The present disclosure provides a method, an apparatus, and a system for collecting an access control list. A second network device receives a first LSA packet flooded by a first network device, where the first LSA packet includes a first network device identifier and first ACL information, and the first network device and the second network device belong to a same IGP area; and sends an extended first BGP-LS packet to a controller, where the extended first BGP-LS packet includes the first network device identifier and the first ACL information, so that the controller can collect ACL information of the first network device and manage the ACL information of the first network device.

Abstract: Embodiments of the present invention provide a label distribution method and device. A first routing device sends a first BGP update message including a first VPN instance identifier and a first RT to a second routing device. The first routing device receives a second BGP update message including a second VPN instance identifier and a second RT. The first routing device allocates a first VPN instance label to a second VPN instance after the second routing device determines that the first RT is equal to the second RT. The first routing device sends a third BGP update message including the first VPN instance label to the second routing device.

Abstract: Embodiments of the present invention provide a Multiprotocol Label Switching traffic engineering tunnel establishing method and device. A tunnel establishing method includes: receiving, by a second routing device, an identifier, which is sent by a first routing device, of an MPLS TE tunnel from a first VPN instance to a second VPN instance; acquiring, by the second routing device according to the identifier, path information of the MPLS TE tunnel from the first VPN instance to the second VPN instance; and establishing an MPLS TE tunnel from the second VPN instance to the first VPN instance according to the acquired path information. Therefore, forward and reverse bidirectional tunnels are co-routed or partially co-routed, thereby solving a problem caused by non-co-routing during BFD.

Abstract: The present application provides a method and a device for clearing a MAC forwarding entry. The method includes: detecting, by a first RB, that a topology of a network accessed by a local terminal changes; and sending, by the first RB, a first packet to a second RB, so that the second RB clears a corresponding forwarding entry after receiving the first packet, where the second RB refers to an RB configured with at least one VLAN the same as that of the first RB.

Abstract: The present application discloses a method, a device and a system for establishing an LSP. The method includes: allocating, by a proxy node device, a label for a destination node device, generating a label mapping message carrying the label, an address of the destination node device and an address of the proxy node device, and sending the label mapping message to an upstream node device to initiate establishment of a first LSP from an entry node device to the proxy node device; stitching, by the proxy node device, the first LSP with a second LSP to form a third LSP from the entry node device to the destination node device, where the second LSP is an LSP established between the proxy node device and the destination node device.

Abstract: Embodiments of the present invention disclose a method, apparatus, and system for detecting performance of a layer 3 virtual private network. A first node creates an OAM instance according to OAM configuration information set by a user, and sends the OAM configuration information to a second nodet; the second node creates an OAM instance according to the OAM configuration information; and the first node performs OAM detection by sending an OAM packet to the second node and/or receiving an OAM packet from the second node, where the sending and/or receiving is implemented by using the OAM instance of the first node. According to the present invention, not only an OAM configuration procedure is simplified and manpower is saved, but also consistency of OAM configuration at two ends can be ensured. In addition, a configuration delay in this dynamic configuration manner may be ignored.

Abstract: Embodiments of the present invention provide a Multiprotocol Label Switching traffic engineering tunnel establishing method and device. A tunnel establishing method includes: receiving, by a second routing device, an identifier, which is sent by a first routing device, of an MPLS TE tunnel from a first VPN instance to a second VPN instance; acquiring, by the second routing device according to the identifier, path information of the MPLS TE tunnel from the first VPN instance to the second VPN instance; and establishing an MPLS TE tunnel from the second VPN instance to the first VPN instance according to the acquired path information. Therefore, forward and reverse bidirectional tunnels are co-routed or partially co-routed, thereby solving a problem caused by non-co-routing during BFD.

Abstract: A first multihoming network virtualization edge (NVE) configured to receive a first multicast packet sent by a multihomed tenant end system (TES), acquire an ingress port of the first multicast packet and a virtual local area network (VLAN) identifier (ID) of the first multicast packet, acquire a virtual overlay network (VN) ID of the first multicast packet according to the ingress port and the VLAN ID, determine whether the ingress port is a designated forwarder (DF) of the VN ID, encapsulate the first multicast packet with an extended NVO3 header when the ingress port is not the DF of the VN ID, and send the first multicast packet that is encapsulated with the extended NVO3 header to another NVE that includes a second multihoming NVE, where the extended NVO3 header carries the VN ID and a link aggregation group (LAG) ID that corresponds to the ingress port.

Abstract: The present application provides a method and a device for clearing a MAC forwarding entry. The method includes: detecting, by a first RB, that a topology of a network accessed by a local terminal changes; and sending, by the first RB, a first packet to a second RB, so that the second RB clears a corresponding forwarding entry after receiving the first packet, where the second RB refers to an RB configured with at least one VLAN the same as that of the first RB. According to embodiments of the present application, after the topology of the network accessed by the local terminal changes, MAC forwarding entries of relevant devices in a whole network can be cleared in time, thereby avoiding a forwarding error.

Abstract: A VPN implementation method and a PE device are provided. The method includes: sending VPN topology information of a local end to a PE device at a peer end, and receiving VPN topology information of the peer end sent by the PE device at the peer end; enabling, according to the VPN topology information of the local end and the VPN topology information of the peer end, the local end and the peer end to select a jointly supported topology; and connecting VPN members by using the jointly supported topology, so as to implement the VPN. A VPN in a multi-topology environment may be implemented, and service quality of a multi-topology VPN may be improved.

Abstract: A Border Gateway Protocol Monitoring Protocol (BMP) module sends to a Border Gateway Protocol (BGP) module a request for acquiring a capability information set; the BGP module traverses BGP capability data stored in the BGP module to acquire all BGP capability data configured in the network device and sends all the BGP capability to the BMP module; and the BMP module sends to the monitoring server an initial capability information set message carrying all the BGP capability data and a device identifier of the network device. By means of capability extension of a network device, the network device can acquire, through interaction between a BGP module and a BMP module, all BGP capabilities configured in the network device and send all the BGP capabilities to a monitoring server, so that the monitoring server can fully understand actual capability supported by the entire network, further providing an effective basis for deployment of and decision on the entire network.

Abstract: Embodiments of the present invention provide a method, a device, and a system for creating a bidirectional multicast distribution tree based on an interior gateway protocol, the method includes: generating, according to a root address of a first node flooded by the first node through an IGP and a shortest path tree algorithm, a shortest path to another node with the first node as a source node, and constructing a flooding tree according to the shortest path; and when receiving layer 3 multicast group information flooded by an edge node, determining, according to the layer 3 multicast group information and a multicast group range that is supported by the first node, a flooding tree corresponding to each layer 3 multicast group, and performing pruning, according to the layer 3 multicast group information, to the flooding tree corresponding to each layer 3 multicast group, to generate a bidirectional multicast distribution tree.

Abstract: Embodiments of the present invention provide a traffic switching method and apparatus. The method includes: establishing a mapping relationship between a performance parameter of a network and a forwarding node switching sequence, where the performance parameter of the network includes link usage before switching and link usage after switching; when traffic switching needs to be performed, obtaining a current performance parameter of the network, and determining, according to the mapping relationship, a switching sequence corresponding to the current performance parameter; and delivering a switching instruction to forwarding nodes according to the switching sequence, so as to control the forwarding nodes to sequentially perform traffic switching. In this way, traffic switching can be systematically implemented, thereby effectively avoiding a link congestion phenomenon in a switching process, and ensuring network performance.