OpenSwitch does not determine the state of a server, even if a connection to the server failed. The server state can be assigned only by an administrator or a coordination module via an RPC such as rp_server_status or within the configuration file.
A server that has a state of UP remains UP, even if the actual server process is no longer running. New connections attempt to establish connections to the server, even though it is actually unavailable. The connections are automatically routed to the next available server after the initial failure.
You can employ a coordination module to have server state established automatically. See the OpenSwitch Coordination Module Reference Manual for instructions on creating and using coordination modules.
