To support high reliability and high availability, modern cloud systems are designed to be resilient to node crashes. That is, the cloud system should gracefully recover from node crashes and continue to function. However, node crashes/reboots that happen at specific time can trigger crash recovery bugs that lie in incorrect crash recovery protocols and their implementations. To ensure that a cloud system is free from crash recovery bugs, various fault injection approaches have been proposed to test whether a cloud system can correctly recover from a variety of fault scenarios. These approaches are not effective enough in exploring the fault scenario space without testers’ knowledge.
In this paper, we present CrashFuzz, a fault injection testing approach that can effectively test recovery code and expose crash recovery bugs of cloud systems. CrashFuzz works by mutating combinations of possible node crashes and reboots according to runtime feedbacks, and prioritizing the combinations that are prone to increase code coverage and trigger crash recovery bugs for smart exploration. We have implemented CrashFuzz and evaluated it on three popular open-source distributed systems including ZooKeeper, HDFS and HBase. CrashFuzz has detected 4 unknown bugs and 1 known bug. Compared with other fault injection approaches, CrashFuzz can detect more bugs.
Fri 19 MayDisplayed time zone: Hobart change
13:45 - 15:15 | Fault injection and mutationJournal-First Papers / NIER - New Ideas and Emerging Results / SEIP - Software Engineering in Practice / DEMO - Demonstrations / Technical Track at Meeting Room 105 Chair(s): Lingxiao Jiang Singapore Management University | ||
13:45 15mTalk | Coverage Guided Fault Injection for Cloud Systems Technical Track Yu Gao Institute of Software, Chinese Academy of Sciences, China, Wensheng Dou Institute of Software Chinese Academy of Sciences, Dong Wang Institute of software, Chinese academy of sciences, Wenhan Feng Institute of Software Chinese Academy of Sciences, Jun Wei Institute of Software at Chinese Academy of Sciences; University of Chinese Academy of Sciences; University of Chinese Academy of Sciences Chongqing School, Hua Zhong Institute of Software Chinese Academy of Sciences, Tao Huang Institute of Software Chinese Academy of Sciences Pre-print | ||
14:00 15mTalk | Diver: Oracle-Guided SMT Solver Testing with Unrestricted Random Mutations Technical Track | ||
14:15 15mTalk | Identifying Defect Injection Risks from Analysis and Design Diagrams: An Industrial Case Study at Sony SEIP - Software Engineering in Practice Yoji Imanishi Sony Global Manufacturing&Operations, Kazuhiro Kumon Sony Global Manufacturing&Operations, Shuji Morisaki Nagoya University | ||
14:30 7mTalk | DaMAT: A Data-driven Mutation Analysis Tool DEMO - Demonstrations Enrico Viganò University of Luxembourg, Oscar Cornejo SnT Centre, University of Luxembourg, Fabrizio Pastore University of Luxembourg, Lionel Briand University of Luxembourg; University of Ottawa Pre-print | ||
14:37 7mTalk | Mutation testing in the wild: findings from GitHub Journal-First Papers Ana B. Sánchez University of Seville, Pedro Delgado-Pérez Universidad de Cádiz, Inmaculada Medina-Bulo Universidad de Cádiz, Sergio Segura University of Seville Link to publication DOI | ||
14:45 7mTalk | An Experimental Assessment of Using Theoretical Defect Predictors to Guide Search-Based Software Testing Journal-First Papers Anjana Perera Oracle Labs, Australia, Aldeida Aleti Monash University, Burak Turhan University of Oulu, Marcel Böhme MPI-SP, Germany and Monash University, Australia Link to publication DOI | ||
14:52 7mTalk | Assurance Cases as Data: A Manifesto NIER - New Ideas and Emerging Results Claudio Menghi McMaster University, Canada, Torin Viger , Alessio Di Sandro University of Toronto, Chris Rees Critical Systems Labs, Jeffrey Joyce Critical System Labs Inc., Marsha Chechik University of Toronto | ||
15:00 7mTalk | Predictive Mutation Analysis via Natural Language Channel in Source Code Journal-First Papers Jinhan Kim KAIST, Juyoung Jeon Handong Global University, Shin Hong Handong Global University, Shin Yoo KAIST Link to publication Pre-print | ||