Method for avoiding routing oscillation in smooth restarting process and smooth restarting equipment

Abstract

The present invention provides a method for avoiding route oscillation in the graceful restart process and a graceful restart device. The fundamental idea of the invention is that, in the graceful restart (GR) process, a GR Restarter leans a corresponding relation between LSA address prefixes and LS ID through the help of a neighbor GR Helper; when the GR process is ended, GR Restarter re-generates the LSA containing address prefixes according to the learned corresponding relation. By utilizing the technical solution provided by the invention, GR process can be realized in the premise of not influencing system performance and reducing realization difficulty, therefore, route oscillation caused by the change of LS ID in the GR process can be avoided.

Description

technical field [0001] The invention relates to a Graceful Restart (GR, Graceful Restart) technology in a routing protocol, in particular to a method for avoiding route oscillation during the Graceful Restart process and a Graceful Restart device. Background technique [0002] At present, most distributed devices on the network adopt a structure in which the control software is separated from the forwarding software. Among them, the control software is responsible for the control and management of the entire device, including running routing protocols for route For the data forwarding service, it maintains its own forwarding table according to the notification message of the control software, and performs data forwarding according to the forwarding table. These two kinds of software are respectively controlled by their respective processors, and are completely independent in function. [0003] GR is a smooth restart technology applied to devices whose control software and f...

AI Technical Summary

Problems solved by technology

If a correspondence relationship between address prefix and LS ID is established for each LSA carrying an address prefix and then backed up to the standby board, the amount of data backed up to the standby board will be very large, which will occupy a lot of precious CPU resources and Inter-board communication channel, which seriously affects system performance

[0023] 2. When route flapping occurs due to unstable network topology, LSAs such as Inter-Area-Prefix-LSA and AS-External-LSA may be quickly deleted and created

In this case, it is required that the backup messages on the standby board must maintain good timing, so as not to process the subsequent backup messages first, resulting in information backup errors

However, since a large number of backup messages will be generated in this process, and it takes a certain amount of time to process these messages, it is difficult to maintain good timing requirements in specific implementation.

Images