In this paper we present some extensions on previously published results regarding connectivity issues in one--dimensional ad--hoc networks. We show how an equivalentGI|D|\infty$ queueing model may be used to address the issue, and present connectivity results on both infinite and finite networks for various node placement statistics. We then show how a GI|G| model may be used to study broadcast percolation problems in ad--hoc networks with general node placement and random communication range. In particular, we obtain explicit results for the case of nodes distributed according to a Poisson distribution operating in a fading environment. In case of nodes distributed according to a Poisson point process, heavy traffic theory is applied to derive the critical communication range for connectivity and the critical transmission power for broadcast percolation in dense networks. The analysis is then extended to the case of unreliable ad--hoc networks, with an in--depth discussion of asymptotic results.