Susceptibility of scale free Power Law (PL) networks to attacks has been traditionally studied in the context of what may be termed as {\em instantaneous attacks}, where a randomly selected set of nodes and edges are deleted while the network is kept {\em static}. In this paper, we shift the focus to the study of {\em progressive} and instantaneous attacks on {\em reactive} grown and random PL networks, which can respond to attacks and take remedial steps. In the process, we present several techniques that managed networks can adopt to minimize the damages during attacks, and also to efficiently recover from the aftermath of successful attacks. For example, we present (i) compensatory dynamics that minimize the damages inflicted by targeted progressive attacks, such as linear-preferential deletions of nodes in grown PL networks; the resulting dynamic naturally leads to the emergence of networks with PL degree distributions with exponential cutoffs; (ii) distributed healing algorithms that can scale the maximum degree of nodes in a PL network using only local decisions, and (iii) efficient means of creating giant connected components in a PL network that has been fragmented by attacks on a large number of high-degree nodes. Such targeted attacks are considered to be a major vulnerability of PL networks; however, our results show that the introduction of only a small number of random edges, through a {\em reverse percolation} process, can restore connectivity, which in turn allows restoration of other topological properties of the original network. Thus, the scale-free nature of the networks can itself be effectively utilized for protection and recovery purposes.