Aging and physical inactivity are two factors that favors the development of cardiovascular disease, metabolic syndrome, obesity, diabetes, and sleep dysfunction. In contrast, the adoption a habitual of moderate exercise may present a non-pharmacological treatment alternative for sleep and metabolic disorders. We aimed to assess the effects of moderate exercise training on sleep quality and on the metabolic profile of elderly people with a sedentary lifestyle. Fourteen male sedentary, healthy, elderly volunteers performed moderate training for 60 minutes/day, 3 days/week for 24 wk at a work rate equivalent to the ventilatory aerobic threshold. The environment was kept at a temperature of 23 ± 2 °C, with an air humidity 60 ± 5%. Blood and polysomnographs analysis were collected 3 times: at baseline (1 week before training began), 3 and 6 months (after 3 and 6 months of training). Training promoted increasing aerobic capacity (relative VO2, time and velocity to VO2max; p < 0.05), and reduced serum NEFA, and insulin concentrations as well as improved HOMA index (p < 0.05), and increased adiponectin levels (p < 0.05), after 3 months of training when compared with baseline data. The sleep parameters, awake time and REM sleep latency were decreased after 6 months exercise training (p < 0.05) in relation baseline values. Our results demonstrate that the moderate exercise training protocol improves the sleep profile in older people, but the metabolism adaptation does not persist. Suggesting that this population requires training strategy modifications as to ensure consistent alterations regarding metabolism.