Business index monitoring method, monitoring server and system

Abstract

The invention discloses a business index monitoring method, being suitable for being executed in a monitoring server. The monitoring server is in communication connection with the time sequence database and the website server. The business index monitoring method comprises the steps of: obtaining a first configuration file of each monitoring item from a website server to generate a corresponding timing monitoring task, wherein the first configuration text at least comprises a monitoring name and a corresponding monitoring expression, and the monitoring expression comprises an alarm condition and at least one index item, and each index item comprises a monitoring index and at least one filtering condition, and a plurality of index items are connected through mathematical operators; when a timing monitoring task of a certain monitoring project is triggered, obtaining a corresponding index value from the time sequence database according to the filtering condition of each monitoring indexin the monitoring expression, and substituting the index value into the monitoring expression to judge whether the monitoring expression is established or not; and if the monitoring expression is established, generating alarm information, and sending an alarm notification. The invention further discloses a corresponding monitoring server and a corresponding monitoring system.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a service indicator monitoring method, a monitoring server and a system. Background technique [0002] With the increase of business systems and more frequent function iterations, the number of problems after the service goes online also increases. How to find problems in the first place becomes the key to the business system. The monitoring of business data will be an effective means to detect online problems in the first place and stop losses in time. However, most of the existing open source monitoring systems focus on operation and maintenance monitoring functions, which are not fully applicable to business monitoring scenarios. More commonly used open source monitoring systems such as Zabbix, Prometheus, Nagios, etc., are usually C / S architecture. C is the client, which is installed on the monitored server and actively collects host indicators and uploads them to the monitor...

AI Technical Summary

Problems solved by technology

For example, the CPU usage rate, memory usage rate, disk space, access to port services and other concerns at the operation and maintenance level, while business data indicators are at a higher level than operation and maintenance indicators, and operation and maintenance monitoring is not applicable

On the other hand, the configuration of monitoring rules is only applicable to the form of a single indicator, which cannot meet the requirements when the data consistency or correlation of multiple indicators needs to be considered for some complex business monitoring

Images