17 fails and 0 passes just isn't a good thing.

I would just look at the tests of the last distribution, or the one I was trying to install. In general, consistent failure like this indicates a poorly written (or just not cpan-tester friendly) test suite.

Update : Guess I should have read the documentation a little more. You need to load the data into the berkeley db before any of the tests pass. The test results could make that a little clearer....

I would disagree with you there. The module should throw a critical error if the required databases don't exist, and the test suite should skip all the tests, although it would be a better idea to just include the data in the distribution.