Failure Categorization for Problem Diagnosis on Exception-Based Software Systems
Traditionally,distributed system software developers print log messages when creating a program to track the runtime status of a system to help identify where problems may have occurred while the program is running.People often use system logs produced by distributed systems for troubleshooting and problem diagnosis.However,there may be thousands of failed jobs occurring within a short time.Manually inspecting these jobs one by one to detect anomalies is unfeasible due to the increasing scale and complexity of distributed systems.Since many fated jobs may have the same cause,there is a great demand for automatic job categorization techniques based on log analysis to help developers prioritize job investigation.Described herein is an unstructured log analysis technique for job categorization.In the technique,we propose a novel algorithm to categorize log messages into different categories without heavily relying on application specific knowledge,based on which jobs can be categorized.
log analysis job clustering message categorization problem diagnosis
Shuhai Li
School of Computer Science and Engineering Beihang University Beijing,China
国际会议
杭州
英文
122-125
2013-03-22（万方平台首次上网日期，不代表论文的发表时间）