A method and device for improving link aggregation protocol timeout

Abstract

The present invention provides a method and device for improving link aggregation protocol timeout, the method comprising the following steps: setting the timeout time and the number of timeout threshold times for the switch to receive the LACPDU packet; in response to the first switch receiving the LACPDU packet from the second switch Overtime, the number of timeouts is counted; in response to the number of timeouts greater than the threshold number of times and the aggregation link of the switch is in the out-of-band management mode, it is judged by the network control center whether the port of the second switch is abnormal; in response to the abnormality of the port of the second switch, the abnormal port of the second switch Remove aggregated links. By using the solution of the present invention, the problem that the stability of the in-band management may be affected by the in-band data flow can be avoided, the stability performance of the aggregation link can be further improved, and the flexibility of forward compatibility is retained.

Description

technical field [0001] This field relates to the computer field, and more particularly to a method and device for improving link aggregation protocol timeout. Background technique [0002] Out-of-band management refers to using an independent management channel for device maintenance. It allows system administrators to remotely monitor and manage servers, routers, network switches and other network devices. In contrast, in-band management refers to using regular data channels to manage devices. An obvious limitation of in-band management (In-Band Management) is that such management is vulnerable to attacks or compromises on managed devices. Out-of-band management addresses this limitation by deploying management channels that are physically separated from data channels. [0003] Aggregation link (Link aggregation) is to aggregate multiple physical ports together to form a logical port to realize the load sharing of outbound and inbound traffic throughput on each member po...

AI Technical Summary

Problems solved by technology

In practical applications, the problem often encountered is that the number of packets that can be sent to the CPU per second is limited on the port to avoid CPU overload, and LACPDU packets need to be sent to the CPU for processing. If the network status is unstable , For example, when a broadcast storm occurs or a loop occurs, there will be too many packets on the port that need to be sent to the CPU for processing, such as ARP packets, which will likely crowd out the LACPDU packets, causing the port to misjudge the port. Remove the aggregated link, so that the port on the aggregated link is in an unstable state

Images