Operation-Tree Differential Debugging for Hint Generation in Network Configuration Exercises

In network-configuration exercises, students often encounter “stumbling blocks” where configuration errors prevent progress. Existing support systems provide limited help: automatic graders restrict flexibility, visualization tools lack explicit diagnosis, and industrial verifiers produce outputs too complex for novices. This study introduces a novel hint generation system that applies differential debugging to network education. The system simulates communication behavior for both a reference configuration and a student's configuration, transforms execution traces into operation trees, and systematically compares them. By detecting the first divergence between the two trees, the method identifies the most likely root cause of failure and generates specific, reproducible hints. These hints highlight the operation involved and the configuration items responsible, while still allowing multiple correct solutions. A prototype implementation in C++ with Graphviz visualization demonstrates the approach. Evaluation shows that hints are generated in about 0.5 seconds for typical student exercises, with scalability to larger topologies. In a field study (N=29), our system achieved higher error-identification success than common network tools; statistically significant improvements were observed in two specific categories—misunderstandings of the overall configuration and detection of embedded errors in a dedicated troubleshooting exercise (Wilcoxon signed-rank, p