A network fault detection method and system

Abstract

The invention discloses a network fault detection method and system. The method is suitable for being executed in an aggregation server; the aggregation server is in communication connection with oneor more network servers and a database server; the network servers are accessed to multiple Internet data centers; each Internet data center is deployed with multiple switches; multiple hosts are accessed to each switch; and a corresponding data probe is preset in each host and is in communication connection with the network server. The method comprises steps: one or more index data reported by the network server are received, wherein the index data are formed by the network server according to a detection result reported by each data probe; according to the index data in an abnormal state, anetwork connection is detected; and if the detection result indicates the network connection fails, an alarm is given; the index data with normal states are subjected to aggregation calculation to acquire corresponding network quality data; and the network quality data are transmitted to the database server for being stored conveniently.

Description

technical field [0001] The invention relates to the field of computer networks, in particular to a network fault detection method and system. Background technique [0002] Internet companies often have tens of thousands of servers. These servers are distributed in multiple computer rooms across the country and are connected to each other to form a company intranet. There are extensive network communications between computer rooms and between servers in the intranet, so the importance of communication quality is self-evident. [0003] A variety of reasons lead to communication quality problems across the computer room intranet, such as network cards, network cables, servers and other single equipment failures. The individual probability of such failures is very small, but after multiplying the huge base, it cannot be ignored, and the single point of failure of the switch, network congestion, etc. , Dedicated line failures, etc. also occur from time to time, so failures are a...

AI Technical Summary

Problems solved by technology

[0003] A variety of reasons lead to communication quality problems across the computer room intranet, such as network cards, network cables, servers and other single equipment failures. The individual probability of such failures is very small, but after multiplying the huge base, it cannot be ignored, and the single point of failure of the switch, network congestion, etc. , Dedicated line failures, etc. also occur frequently, so failures are almost unavoidable, so it is necessary to monitor the network quality across computer rooms and locate faults in a timely and accurate manner, which is very important to ensure the normal operation of the company's business

[0004] Existing network fault detection solutions are mostly divided into two categories. One is to manually locate the fault. After the problem has occurred, the operation and maintenance engineer will gradually narrow down the scope based on experience to troubleshoot suspicious lines, switches or servers. It is time-consuming and laborious and lacks data support. The accuracy is low, and the other is to actively detect the network status by deploying probes on all online servers, but there are problems that the network topology coverage is incomplete and the detection data is difficult to comprehensively utilize

Images