The Study of Container-oriented Deployment Automation for Vulnerability Scanning Systems

• Main Authors: Tsao Chung-Chieh, 曹仲杰
• Other Authors: Li Wen-Ting
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/yc3746

碩士 === 國立高雄師範大學 === 軟體工程與管理學系 === 107 === Software deployment takes a lot of time to manually adjust the parameters, control remote server, install kits, set default account, password or other interactive information that requires user input to complete the setting and installation of each computer. This paper applies the concept of development and operations (DevOps) to automate the deployment process. Deployment automation needs to overcome the two problems, automatic interactive input and automatic parameter replacement. This study provides an automated deployment architecture, security information system installation process, and automatic interactive input approach. The compiled shell scripts and Python programs are imported into Jenkins to achieve simultaneous deployment and verification of multiple software systems automatically. This study compares the time spent on automated and manual deployments of OpenVAS, Nagios and Arachni security systems. The performance of Arachni system deployment and scanning between virtual machines and virtual containers are also compared. The result shows that using the provided automated deployment solutions can save a lot of deployment time and reduce the burden of technical staffs while establishing basic system environments and maintaining daily operations.