We have lots of test failures. It is not efficient to track them in individual issues. This one tracks them all, until we are near-clean.