On Sat, Dec 01, 2012, Ross Walker wrote:
>
> But isn't that a good thing? If someone submits something to the tree that
> leads to one of the parallel tests hanging we want the commit marked as a
> failure and it to remain that way until the problem is rectified. If it
> just skips a hung test and carries on running the rest of the tests then
> it is not immediately obvious that somebody broke something.
Why is a bug that leads to an MPI hang more important to mark than some other
(equally bad) error where the job finishes quickly?
Second, presumably a test that failed because of a new "MPI timeout" would
still count as a failed test. But cruise control would still be able to
go on to other tests in the test suite, so one can tell if one has a isolated
failure or a generic one.
Both these arguments suggest that is is a "good thing" to limit the time spent
on any individual test.
...dac
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Sat Dec 01 2012 - 17:30:02 PST
