Treating bugs as allergies: a safe method for surviving software failures
作者： Yuanyuan Zhou , Feng Qin , Joseph Tucek
DOI:
关键词:
摘要: Many applications demand availability. Unfortunately, software failures greatly reduce system Previous approaches for surviving suffer from several limitations, including requiring application restructuring, failing to address deterministic bugs, unsafely speculating on program execution, and a long recovery time. This paper proposes an innovative, safe technique, called Rx, that can quickly recover programs many types of common both non-deterministic. Our idea, inspired by allergy treatment in real life, is rollback the recent checkpoint upon failure, then reexecute modified environment. We base this idea observation bugs are correlated with execution environment, therefore be avoided removing "allergen" Rx requires few no modifications provides programmers additional feedback bug diagnosis.
参考文章(27)
Srikanth Kandula, Yuanyuan Zhou, Sudarshan M. Srinivasan, Christopher R. Andrews, Flashback: a lightweight extension for rollback and deterministic replay for software debugging usenix annual technical conference. pp. 3- 3 ,(2004)
Srikanth Kandula, Yuanyuan Zhou, Sudarshan Srinivasan, Christopher Andrews, Flashback: A Light-weight Rollback and Deterministic Replay Extension for Software Debugging ,(2004)
W. Vogels, The Design and Architecture of the Microsoft Cluster Service ieee international symposium on fault tolerant computing. pp. 422- 431 ,(1998)
Michael E. Locasto, Angelos D. Keromytis, Stelios Sidiroglou, Stephen W. Boyd, Building a reactive immune system for software services usenix annual technical conference. pp. 11- 11 ,(2005) , 10.7916/D86D6562
Martin Rinard, Cristian Cadar, William S. Beebee, Daniel M. Roy, Tudor Leu, Daniel Dumitran, Enhancing server availability and security through failure-oblivious computing operating systems design and implementation. pp. 21- 21 ,(2004)
Hal Stern, Evan Marcus, Blueprints for high availability ,(2000)
Hal Stern, Evan Marcus, Blueprints for high availability: designing resilient distributed systems John Wiley & Sons, Inc.. ,(2000)
Perry Wagle, Jonathan Walpole, Calton Pu, Steve Beattie, Aaron Grier, Crispin Cowan, Heather Hintony, Qian Zhang, Peat Bakke, Dave Maier, StackGuard: automatic adaptive detection and prevention of buffer-overflow attacks usenix security symposium. pp. 5- 5 ,(1998)
M. Rinard, C. Cadar, D. Dumitran, D.M. Roy, Tudor Leu, A dynamic technique for eliminating buffer overflow vulnerabilities (and other memory errors) annual computer security applications conference. pp. 82- 90 ,(2004) , 10.1109/CSAC.2004.2
Teck Chia, Werner Vogels, Katherine Guo, Dan Dumitriu, Ashutosh Agrawal, Scalability of the microsoft cluster service conference on usenix windows nt symposium. pp. 2- 2 ,(1998)