Computer Science Department
McGregory Hall, 3rd Floor
13 Oak Drive
Hamilton, NY 13346
Charlotte Jablonski, Administrative Assistant
Modern computer networks must satisfy a complex and evolving set of security, availability, and performance requirements to meet the diverse needs of users and applications. Configuring a network's routers to satisfy these policies is a daunting task due to the low-level of abstraction exposed by a router's control plane, the intertwined nature of requirements across traffic classes, and the desire to maintain policy-compliance under a range of network failures. Consequently, router configurations are highly prone to bugs that often lead to security vulnerabilities and network outages.
To address this issue, Prof. Aaron Gember-Jacobson, along with collaborators from the University of Wisconsin-Madison, Intentionet, and Microsoft Research, have created a system, called CPR (for Control Plane Repair), that automatically repairs network configuration errors. Inspired by recent work in automated program repair, CPR casts configuration repair as a maximum satisfiability modulo theory (MaxSMT) problem whose constraints are
based on a digraph-based representation of a network’s semantics. Crucially, this representation captures the dependencies between traffic classes arising from the cross-traffic-class nature of control plane constructs. The MaxSMT formulation accounts for these dependencies whilst also accounting for all policies and preferring repairs that minimize the size (e.g., number of lines) of the configuration changes. Using configurations from 96 data center networks, we have shown that CPR produces repairs in less than a minute for 98% of the networks, and these repairs requiring changing the same or fewer lines of configuration than hand-written repairs in 79% of cases.
The full paper is available at: http://aaron.gember-jacobson.com/docs/gember-jacobson2017cpr.pdf