There are tests involving random number generation where the test fails time to time due to stochastic reasons, with relatively small probability (about 10% in the worst case). It would be an improvement to automatically re-run on failure these tests once again.
There are tests involving random number generation where the test fails time to time due to stochastic reasons, with relatively small probability (about 10% in the worst case). It would be an improvement to automatically re-run on failure these tests once again.