An intelligent root cause positioning method and system for IT faults based on multi-source log analysis

CN122309213APending Publication Date: 2026-06-30HANGZHOU ZHISHUCHANGLIAN TECH CO LTD

Patent Information

Authority / Receiving Office: CN · China
Patent Type: Applications(China)
Current Assignee / Owner: HANGZHOU ZHISHUCHANGLIAN TECH CO LTD
Filing Date: 2026-04-07
Publication Date: 2026-06-30

AI Technical Summary

Technical Problem

Existing IT fault location methods rely on a single data source, which requires operations and maintenance personnel to manually switch between multiple interfaces, resulting in low location efficiency and an inability to quickly and accurately locate the root cause of the fault from multi-source operations and maintenance data.

Method used

A multi-source log analysis method is adopted to collect log, metric and call chain data, build an abnormal event library, generate a candidate root cause event list through time axis alignment and collision retrieval, and combine call chain data for pruning and context analysis to locate the root cause of the failure.

Benefits of technology

It enables the automatic generation of a list of suspected events sorted by time within a fault time window, allowing maintenance personnel to quickly pinpoint the root cause of a fault without switching between multiple systems, thus improving the efficiency of IT fault root cause localization.

✦ Generated by Eureka AI based on patent content.

Smart Images

Figure CN122309213A_ABST

Patent Text Reader

Abstract

This invention relates to the field of fault root cause localization methods, specifically to an intelligent IT fault root cause localization method and system based on multi-source log analysis. The method includes: collecting multi-source operation and maintenance data; processing the multi-source operation and maintenance data in real time; extracting abnormal events; and constructing an abnormal event database. In response to the detection of a business fault, a fault time window is defined. Within the fault time window, abnormal events in the abnormal event database are aligned on a timeline and subjected to collision detection to generate a candidate root cause event list sorted by time. Based on the candidate root cause event list, pruning and contextual analysis are performed using call chain data to locate the fault root cause. Using this method, operation and maintenance personnel can intuitively see the time sequence of key events before and after a fault without manually switching between multiple systems, quickly pinpointing the most likely root cause and improving the efficiency of IT fault root cause localization.

Need to check novelty before this filing date? Find Prior Art