The Analysis And Optimization For High Availability Fault Management Mechanism Of KYLIN Operating System

The paper is proposed to analyze and optimize the high availability fault management system combined software and hardware together,make the KYLIN ope ration system as the basis with the ke y po int of fault management after the study,research and analysis of current popular operation systems,such as Linux,Unix,Windows and so on.The main po ints are as follows:1.Aiming at the fault management mechanism of mainstream ope rating systems such as Linux,Unix,Windows,and so on,finished the related analysis and research about the unified description mechanism of fault format and unified algorithm of time series,MCA / EDAC / AER,the mechanism of Eventing Tracing,perf tool,fault response and failover.2.Due to the deficiency of fault management of current operation system,designed a unified fault management frame of single machine based on the failover system of Event Tracing and proposed a high availability fault management system of multi-machines.3.Using the active monitor mode and ABRT fault management tool to analyze and optimize existing monitoring mode of operating system.Including hardware error report,Daemon operation mode,auto fault report system and so on.4.Select 6 classical faults to test the active monitor method in application layer and internal core layer.At the same time,carry out the CGL standard test to the availability of computer system.In a conclus ion,the thesis foc us on the de ficiency of current ope ration system,for example,single-machine management,variety types and fragmented too ls and so on and proposed a high availability fault management system.It is very practical to modify current fault test mode and make the high availability fault management more reasonable and integrated.