netroub: Towards an Emulation Platform for Network Trouble Scenarios
Colin Regal-Mezin, Satoru Kobayashi, Toshihiro Yamauchi · 2023
Network outage data is valuable and difficult to use, so it is helpful for operators to have a tool that facilitates both the generation of synthetic data and its use. In this paper, we propose the design and implementation of netroub, a platform capable of emulating complex and realistic network trouble scenarios to collect synthetic operational data. Netroub reproduces network failure behaviors described in a scenario file on Docker-based emulation networks. We preliminarily evaluate our platform in two ways: by observing how the emulated failure behaviors increase the diversity of logs through twenty scenarios, and by demonstrating the correct functioning of netroub through a case study.