Log-Based Cloud Monitoring System for OpenStack

摘要

Cloud systems are fast emerging remote computing environment, providing their users with computational resources and services on demand over the Internet. OpenStack is one representative open source platform which helps in building private clouds for the organization. Monitoring the activities in the cloud is a task of utmost importance and log files are often used for it. Logs management plays a very important role in determining various issues in the cloud network or helps in troubleshooting the issues. Often logs are cumbersome to manage and difficult to read. In this paper, we propose a system which monitor the logs of OpenStack components (i.e., Nova, Neutron, Cinder) in real-time and generate an alert for the information, debug, error, warning, and trace messages as soon as an issue recorded in the logs. It helps in easy readability to the administrators. Apart from that our logic also helps to generate warning messages prior to the different resource quota like virtual CPUs, Floating IP's, etc. A feature like that can help the administrator to increase the quota or release some and avoid the inconvenience caused to the users creates virtual machines. We develop our proposed system and demonstrate its feasibility and performances through simulation.