Recovering from intrusions in distributed systems with DARE
Taesoo Kim, Ramesh Chandra, Nickolai Zeldovich · 2012
Dare is a system that recovers system integrity after intrusions that spread between machines in a distributed system. Dare extends the rollback-and-reexecute recovery model of Retro [14] to distributed system recovery by solving the following challenges: tracking dependencies across machines, repairing network connections, minimizing distributed repair, and dealing with long-running daemon processes. This paper describes an early prototype of Dare, presents some preliminary results, and discusses open problems.