As you know, the GMT has a large set of tests. Run these tests - the task is not quite trivial, but possible. A number of tests ended with an error. Do you need data on these tests and in what form?
fail_count.d (1.41 KB)
Thanks, we run these tests all the time here. About 18 is failing for me - they are there to remind us of things that are nontrivial to fix.