A system and method for uninterrupted routing during master-slave disk switching of control plane

Abstract

The invention discloses a system and method for uninterrupted routing during master and backup disk switching of the control plane, and relates to the field of communication technology. In the socket proxy module, the stored socket information is then synchronized to the second socket proxy module; after receiving the socket information, the second socket proxy module stores the socket information and calls the standby The TCP / IP module interface of the disk obtains the corresponding socket; when the active disk and the standby disk are switched, the second socket proxy module traverses the socket information stored locally, and replaces the standby disk protocol module to provide information related to each socket. The keep-alive message of the corresponding protocol; after the switchover between the active disk and the standby disk is completed, the socket agent module stops replacing the standby disk protocol module to provide the keep-alive message. The invention can keep the connection state with the neighbor device node protocol module undisconnected during the promotion process of the spare disk, and realize uninterrupted routing.

Description

technical field [0001] The invention relates to the technical field of communications, in particular to a system and method for uninterrupted routing during master and backup disk switching of a control plane. Background technique [0002] For carrier-class routers, non-stop routing technology (NSR, Non-Stop Routing) is one of the core technologies to achieve five nines or more reliability, and non-stop routing technology is a threshold-level requirement related to the reliability of high-end routers. Compared with non-high-availability or standardized NSF (Non-Stop Forwarding, non-stop forwarding) technology, NSR can confine network failures to nodes, and ensure that the data plane does not experience traffic interruption when the control plane fails. Higher reliability; and compared with NSF technology, its deployment environment requirements are lower, and it does not require the cooperation of neighboring nodes. [0003] In general, in order to achieve uninterrupted rou...

AI Technical Summary

Problems solved by technology

Depending on the control plane protocol stack used, the upgrade process from the standby disk to the successful sending of keep-alive messages may take seconds or even minutes.

[0004] In particular, in the routing device in the application scenario where the backup disk program is inactive or partially activated, that is, the backup disk module only has the data synchronization function with the active disk module, and the state machine of the protocol module does not migrate normally , the upgrade process may take more than ten minutes. During this time period, the standby disk protocol module does not have the ability to create and send messages, and cannot send keep-alive messages.

Since the keep-alive time of the keep-alive packets in common protocol modules is generally within 1 minute, if the keep-alive time is exceeded, the protocol status between the data plane and neighboring device nodes will become DOWN, which will cause the data plane routing to be withdrawn, resulting in traffic Emergence of severe consequences of disruption

Images