NVision-PA: A Tool for Visual Analysis of Command Behavior Based on Process Accounting Logs (with a Case Study in HPC Cluster Security)

Abstract

In the UNIX/Linux environment the kernel can log every command process created by every user with process accounting. Thus process accounting logs have many potential uses, particularly the monitoring and forensic investigation of security events. Previous work successfully leveraged the use of process accounting logs to identify a difficult to detect and damaging intrusion against high performance computing (HPC) clusters, masquerade attacks, where intruders masquerade as legitimate users with purloined authentication credentials. While masqueraders on HPC clusters were found to be identifiable with a high accuracy (greater than 90%), this accuracy is still not high enough for HPC production environments where greater than 99% accuracy is needed. This paper incrementally advances the goal of more accurately identifying masqueraders on HPC clusters by seeking to identify features within command sets that distinguish masqueraders. To accomplish this goal, we created NVision-PA, a software tool that produces text and graphic statistical summaries describing input processing accounting logs. We report NVision-PA results describing two different process accounting logs; one from Internet usage and one from HPC cluster usage. These results identify the distinguishing features of Internet users (as proxies for masqueraders) posing as clusters users. This research is both a promising next step toward creating a real-time masquerade detection sensor for production HPC clusters as well as providing another tool for system administrators to use for statistically monitoring and managing legitimate workloads (as indicated by command usage) in HPC environments.