Debugging Network Management Scripting Applications

A network management scripting application consists of a set of distributed and possibly mobile delegated management scripts [1, 2], as well as other non-script network management processes, which cooperate with each other to accomplish a given network management task. Such applications require debugging support: rstly, they are distributed applications, and distributed applications are notoriously di cult to debug. More importantly, the network management and the scripting environment exhibit many peculiar characteristics which necessitate special debugging support. The overall issue addressed by this paper is: how can we e ectively debug distributed and mobile network management scripting applications? Race conditions and non-deterministic behavior in scripting applications make the cyclical debugging technique not directly applicable. Record-replay debugging solves this problem by deterministically \replaying" an execution using the information gathered in the initial execution. Of the two variations of this approach, simulation replay and instant replay, we found the former to be more suitable to our environment because it does not need the full participation and control of all the processes in the application. Another very powerful technique, global property-based predicates, allows questions to be asked about the distributed computations based on local and global states to help understanding the computations. One di cult problem that arises in replay due to a peculiar characteristic of network management scripting applications is that it is not always possible to synchronize the replay of the scripts. This happens when a synchronization is through a non-script process because of which we are unable to observe and replay it. Possible approaches to this intrinsic problem include simply ignoring the problem by not providing replay synchronization, introducing physical clocks which is not an easy task by itself, and putting the burden on the user Prepared through collaborative participation in the Advanced Telecommunications/Information Distribution Research Program (ATIRP) Consortium sponsored by the U.S. Army Research Laboratory under the Federated Laboratory Program, Cooperative Agreement DAAL01-96-2-0002. This work was also supported in part by University of Delaware 1998-1999 academic year University Competitive Fellowship.