Hadza (along with Sandawe) are the two languages with click consonants not spoken in Namibia, South Africa or Botswana. It is spoken in Tanzania by an uncertain (and probably dwindling) number of people around Lake Victoria. Ethnologue has the language use as “vigorous” but most of the web sites that I looked at seemed to portray the current situation as pretty bad.
There are a lot of very interesting classification issues with these languages. Greenberg classified Hadza as Khoi-San, but the basis for the classification seems to have been “all obviously non-Bantu click languages” rather than positive evidence of shared material. Since then, however, there’s been quite a bit of work which shows that there is a single family of click languages. Bonny Sands, for example, has a number of papers about this, and there’s a nice paper by Tom Gueldemann on the morphological evidence.
The Wikipedia page on the language has the phoneme inventory and a couple of links to external bibliographies. You can hear sound files of Hadza from the UCLA phonetics lab. A link to the page is here.
Helen Eaton has written a couple of papers on different aspects of Hadza grammar, including this paper on focus. There’s also this paper by Niklas Edenmyr on the semantics of gender assignment. Apart from this, a great deal of the Web presence of Hadza seems to be “reprints” of the Wikipedia article.