A method, system and computer-readable storage medium for determining the root cause of a fault

Abstract

The embodiment of the present invention discloses a method, system and computer-readable storage medium for determining the root cause of a fault, which analyzes the fault source information from three dimensions of multi-sequence correlation, data stream correlation and event correlation, and obtains the first fault index Abnormal index set; use the Pearson-gray correlation combination method to calculate the degree of correlation between the first fault index and each abnormal index in the corresponding abnormal index set; use the abnormal index corresponding to the correlation degree that meets the preset threshold as the association Result set; query the pre-established fault diagnosis set to obtain the historical diagnosis result set corresponding to the first fault index; calculate the intersection of the association result set and the historical diagnosis result set, and the index contained in the intersection is the fault of the first fault index Root cause. According to this technical solution, the automatic and intelligent mining and locating of the root causes of faults in the e-commerce system are realized, and the speed of problem solving by development and operation and maintenance personnel is effectively improved.

Description

technical field [0001] The present invention relates to the technical field of fault analysis, in particular to a method, system and computer-readable storage medium for determining the root cause of a fault. Background technique [0002] With the expansion of business and monitoring scale, the monitoring indicators of e-commerce system are increasing day by day. At the same time, when monitoring indicators fail, it is difficult for development and operation and maintenance engineers to quickly find the root cause of the failure from the massive monitoring indicators. The "alarm storm" greatly interferes with the engineer's thinking of locating the problem; the speed of fault recovery basically depends on the experience and operation speed of the engineer. At present, the industry mainly uses monitoring systems to monitor and collect e-commerce systems in real time, detect abnormalities, and provide fault alarms. [0003] In the face of complex business monitoring and prob...

AI Technical Summary

Problems solved by technology

This method is inefficient through manual investigation, and it takes a long time to analyze and locate the cause of the fault. It is even more difficult to dig out the hidden root cause of the fault.

[0004] A complete monitoring closed loop should include: monitoring collection, abnormal detection, fault alarm, problem location and automatic resolution. The monitoring system in the industry mainly focuses on anomaly detection and fault alarm, and there is still a lack of comprehensive analysis and positioning of faults in e-commerce systems. A complete set of methods and systems

Images