Service robotics in dynamic environments is a challenging yet exciting field that has witnessed incredible advances in the last few decades. The credit for these advancements goes to artificial intelligence and robotics technological progressions. The housekeeping domain is an example of a service robotics domain. A high-level task planner can compute a plan of action, while low-level functionalities like perception and motion planning can ensure feasibility. We propose a solution to the housekeeping problem that utilizes the expressiveness of action language C+ to formalize actions and compute an action plan. To check for feasibility in the real world, we execute the action plans in a physics simulator, GAZEBO. We also propose benchmark instances for the housekeeping domain and evaluate our planning and execution times approach. We report possible execution failures for the benchmark instances and offer plausible solutions, such as hybrid planning, plan execution monitoring, and conditional planning.