The Design And Implementation Of LAN Fault Management Rule Engine

With the increasing complexity of LAN layout and topology,the failure frequency of LAN is increasing.In order to ensure the reliability of LAN operation,it is necessary to establish LAN fault management system to manage network faults efficiently.According to the actual deployment and topology of LAN,the traditional LAN fault management scheme hard codes the network fault detection logic in the form of program code in the network fault management system.There are some defects in such a scheme.First of all,because the LAN equipment and topology will continue to change,resulting in each change,developers need to directly modify the system source code,which will cost a lot.Secondly,due to the complexity of LAN topology,the logic judgment of network fault detection is also very complex.A large number of logic judgment program codes are coupled together,which makes the maintenance of the system very poor.Using rule engine technology for LAN fault management can effectively solve the above problems.Firstly,the rule engine decouples the logic code of network fault detection from the system,describes the network fault detection logic with rules in specific data format,such as JSON,XML,etc.Every time the LAN changes,developers do not need to modify the system source code,only the corresponding rules can be modified.Secondly,for a local area network,the network fault detection logic can be defined as multiple rules,which improves the scalability and maintainability of the system.The thesis describes the solution of LAN fault management rule engine proposed by a domestic W company,and describes the design and implementation of the core module of the engine in detail.The rule engine uses rules to describe the detection logic of network fault.The fault alarm reported in the network will trigger the execution of the corresponding rules.Through the execution of the rules,the alarm source is located,and the fault causes and solutions are discovered.The engine consists of four core modules: design module,asset management module,rule parsing module and rule execution module.Network operation and maintenance experts design rules through design module,asset management module manages rule information,rule parsing module parses the activated rules,and rule execution module obtains alarm messages for rule execution.The rule engine is based on W company's Paa S platform Fusion Stage 2.0,which is developed by using W company's Micro service framework CSE.The front and back platforms use the company's self-developed framework BME,and connect with Kafka component consumption alarm events.Zoo Keeper is used for cluster management to ensure the reliability of the rule engine.At present,the rule engine has been put into use in China Unicom,China Telecom and other companies,effectively improving the efficiency of LAN fault management.