OSDI '04 Abstract
Pp. 245258 of the Proceedings
Automatic Misconfiguration Troubleshooting with PeerPressure
Helen J. Wang, John C. Platt, Yu Chen, Ruyun Zhang, and Yi-Min Wang, Microsoft Research
Technical support contributes 17% of the total cost of ownership of
today's desktop PCs. An important element
of technical support is troubleshooting misconfigured applications.
Misconfiguration troubleshooting is particularly challenging, because
configuration information is shared and altered by multiple
In this paper, we present a novel troubleshooting system:
PeerPressure, which uses statistics from a set of sample machines
to diagnose the root-cause misconfigurations on a
sick machine. This is in contrast with methods that require manual
identification on a healthy machine for diagnosing
misconfigurations. The elimination of this manual
operation makes a significant step towards automated misconfiguration
In PeerPressure, we introduce a ranking metric for misconfiguration
candidates. This metric is based on empirical Bayesian estimation. We
have prototyped a PeerPressure troubleshooting system and used a database of
87 machine configuration snapshots to evaluate its performance. With 20
real-world troubleshooting cases, PeerPressure can effectively pinpoint the
root-cause misconfigurations for 12 of these cases. For the remaining cases,
PeerPressure significantly narrows down the number of root-cause candidates
by three orders of magnitude.
- View the full text of this paper in HTML and
Until December 2005, you will need your USENIX membership identification in order to access the full papers. The Proceedings are published as a collective work, © 2004 by the USENIX Association. All Rights Reserved. Rights to individual papers remain with the author or the author's employer. Permission is granted for the noncommercial reproduction of the complete work for educational or research purposes. USENIX acknowledges all trademarks within this paper.
- If you need the latest Adobe Acrobat Reader, you can download it from Adobe's site.