Many tests require random numbers in order to effectively test. Normal runs will need random seeds to get good test coverage, but it must be possible to reproduce such a run. As such logs need to save the root random seed and provide some easy way of reusing that seed for debugging.
Many tests require random numbers in order to effectively test. Normal runs will need random seeds to get good test coverage, but it must be possible to reproduce such a run. As such logs need to save the root random seed and provide some easy way of reusing that seed for debugging.