Abstract:
The embodiments of the present invention provide a tunnel fault detection method and a traffic engineering (TE) node. The method includes: receiving, by a traffic engineering TE node, a link fault report packet; determining, by the node, a faulty link in which a fault occurs according to the link fault report packet; detecting, by the node, whether one or more tunnels which use the node as an ingress node and pass through the faulty link exist; and identifying the tunnels as faulty tunnels if the one or more tunnels which use the node as the ingress node and pass through the faulty link exist. In the embodiments of the present invention, the TE node can detect all faulty tunnels at a time, and it is not necessary to wait until the fault notification based on each tunnel is received and then the faulty tunnel is identified one by one.
Abstract:
Embodiments of the present invention provide a method and an apparatus for forwarding multicast traffic. The method includes: receiving a third multicast join message; in response to the third multicast join message, sending a first multicast join message to a first upstream router, and establishing an active path; in response to the third multicast join message, sending a second multicast join message to a second upstream router, and establishing a standby path; and sending multicast traffic to the multicast receiver through the active path, where the standby path does not forward the multicast traffic. In the foregoing embodiments, the standby path that does not forward the multicast traffic is established beforehand, so that when a fault occurs in the active path, the standby path established beforehand can be used to forward the multicast traffic.
Abstract:
Embodiments of the present invention provide a method and an apparatus for forwarding multicast traffic. The method includes: receiving a third multicast join message; in response to the third multicast join message, sending a first multicast join message to a first upstream router, and establishing an active path; in response to the third multicast join message, sending a second multicast join message to a second upstream router, and establishing a standby path; and sending multicast traffic to the multicast receiver through the active path, where the standby path does not forward the multicast traffic. In the foregoing embodiments, the standby path that does not forward the multicast traffic is established beforehand, so that when a fault occurs in the active path, the standby path established beforehand can be used to forward the multicast traffic.
Abstract:
Embodiments of the present application disclose a multipath transmission based packet traffic control method, including: monitoring egress traffic of a data packet set sent by a source node; determining, according to a preconfigured mapping relationship between traffic ranges and forwarding paths, a target forwarding path set corresponding to the egress traffic of the data packet set, and a weight parameter of the egress traffic borne on each target forwarding path in the target forwarding path set; and distributing the data packet set according to the weight parameter, and transferring the data packet set to a destination node by the target forwarding path set. The embodiments of the present application further disclose a traffic control apparatus. By using the present application, centralized policing and classification can be performed on user traffic in a multipath transmission scenario.
Abstract:
The present invention provides a control channel establishing method, a forwarding point, and a controller. The method includes: sending, by a first FP, topology information of the first FP to a second FP by using the LLDP; receiving, by the first FP, first routing information that is sent by the controller and is used by the first FP to reach the controller, where the first routing information is generated by the controller according to the topology information of the first FP; and establishing, by the first FP, a second control channel with the controller according to the first routing information. The first FP does not need to run the IGP protocol, thereby avoiding that the first FP maintains an IGP neighbor relationship; moreover, the first FP does not need to store routing information for reaching another FP, which helps save storage resources and helps improve forwarding performance.