All modern control systems (CS’s) of large experimental facilities are distributed. Nodes are connected via Ethernet or via specific links such as CAN, RS232/485, GPIB, etc. During normal operation, all nodes are online and connections between them are established and work. However, during operation some nodes can become unavailable for various reasons: electronic failure, power-supply failure, communication line break, human error, etc. Therefore, the task of restoring connections between nodes arises (after the problems are solved). In this paper, the implementation of automatic reconnects in various situations is discussed based on the experience of the VEPP-5 Injection Complex CS at the Budker Institute of Nuclear Physics, Siberian Branch, Russian Academy of Sciences.