Research And Implementation Of Intelligent Ops System Based On Docker

With the popularity of the Internet agile development model,the software development cycle is shortened,the iteration speed is faster,and the traditional development and operation and maintenance methods are less efficient.After the developers complete the specified functions,they need to manually integrate the code of each module.When the code is integrated,they need to be compiled and deployed manually.However,the way that relies on manual operation is prone to errors,and when errors occur,developers need to spend extra time and effort to deal with the problem,resulting in project delays.After the code is deployed and launched,if an error occurs within the service,the developers cannot quickly perceive it.In addition,user access traffic is uncontrollable.During traffic peaks,service resources may not be able to support high concurrent access.During traffic troughs,redundant service resources will increase enterprise costs.During the operation of online services,if you need to fix code problems or add new functions,you need to stop the service first,which will affect user experience and reduce corporate profits.Based on the above problems,this paper proposes and implements an intelligent Ops system.Our solution combines several new technique in software engineering together and achieves efficient developing and operating.First,we use Jenkins and Git Lab to achieve continuous integration and building functions;Second,we use Dockerfile to create images,generates Docker containers from images,and then uses automated Bash scripts to push them to the Kubernetes cluster to achieve automated deployment;Third,we use Log Tail data collection technology to collect data,and then uses Grafana The technology visualizes the data and provides the monitoring and alarm function within the service;Fourth,the elastic scaling algorithm is used to realize the elastic scaling function to avoid the problem of insufficient service resources affecting the online service;Finally,the Erlang hotfix technology is used to realize the hot loading function of the online service.This system adopts the front-end and back-end separation architecture.The front-end part uses Iddas technology for permission verification,and uses Vue.js and Phoniex framework to render the interface;the back-end system adopts MVC architecture and uses OTP framework for development.The main functions of the system include authority and account management,automatic continuous integration,automatic compilation,automatic deployment,monitoring and alarming,automatic elastic scaling,hotfix,etc.We have conducted many rounds of test and the results show that our system has good stability.The system has been connected to the company's live chat project,and after many adjustments and iterations,it has achieved good user feedback.This system can completely liberate the R?D personnel from the traditional development and operation and maintenance methods,so that they can focus on product R?D,and finally effectively improve the R?D efficiency and product quality,and have good actual economic benefits.