Low-Power and Lossy Network (LLN) is composed of embedded devices with limited power, memory, and processing resources. LLN has a wide variety of applications including industrial monitoring, connected home, health care, urban sensor networks and environmental monitoring. LLN uses Routing Protocol for Low-power and lossy networks (RPL) protocol. The RPL maintains directed acyclic graphs for routing packets. By exploiting some features, a Distributed Denial-of-Service (DDoS) attack can be conducted easily. DDoS attacks are very popular and well studied in the context of the Internet, but not in the context of LLNs. In this paper, we propose a powerful DDoS attack framework in LLNs. We formulate the attack as an optimization problem for selecting an optimal set of attackers and their targeted neighbors constrained by a limited link bandwidth. We propose an optimal solution by transforming the optimization problem into a max-flow problem. We provide simulations to support our model.