Takuan: Using Dynamic Invariants to Debug Order-Dependent Flaky Tests
Nate Levin, Chengpeng Li, Yule Zhang, August Shi, Wing Lam · 2025
Automated regression testing is critical to effective software development, but it suffers from flaky tests, i.e., tests that can nondeterministically pass or fail when run on the same version of code. Conceptually, a flaky test depends on a component not controlled by the test, and the test’s outcome depends on the state of the component. For example, a prominent type of flaky tests is order-dependent (OD) tests, whose results depend on the unspecified order in which they are run, as a result of some other test "polluting" shared state. We propose the use of dynamic invariants to help debug flaky tests. By capturing the dynamic invariants that hold true during a passing run and comparing them against those captured during a failing run, we can isolate the reason for the flaky behavior.To illustrate the potential of using dynamic invariants for this task, we implement Takuan, a technique for debugging OD tests by analyzing differences in dynamic invariants collected in the passing and failing runs of OD tests. Invariants that hold true in a passing order but not in a failing order can indicate the "clean" value of the shared state that makes the test pass. To demonstrate how these invariants can be used to debug and repair OD tests, we develop automated approaches that use the invariants to search for methods that can reset the shared state back to the "clean" state. Takuan’s ability to analyze polluted external shared state (e.g., in the file system) allows it to handle cases that prior work cannot. We conduct a preliminary study of Takuan on existing OD tests; Takuan provides an average runtime improvement of 88.1% over prior work while handling more OD tests.