I am just looking for a little clarification here. From my instructions I have identified 4 modes that application can be started.
1. Start in standalone mode. The application bypasses the networking entirely and everything runs on the one machine. In this instance the user is presented with a configuration screen at startup that allows them to configure location of the database only. This box has been prepopulated with the db location from suncerify.properties. 2. Start in server mode. This will start the RMI server only. Presumably the examiner is then going to come along and start the client (option 3) on a different machine to prove the networking works. The user is presented with a configuration screen at startup that allows them to configure location of the database only. This box has been prepopulated with the db location from suncerify.properties. 3. Start with no arguments. This assumes that there is an RMI server running somewhere and the user is presented with a configuration screen that allows them to configure the host and port that they can connect to. Since the db access is on the server side you cannot connect to a different db when starting the client. 4. Start with invalid arguments. This just throws and exception and tells the user they have not used valid arguments.
So my question: I just want to make sure that my understanding of the networking stuff is correct. When they start in server mode they are just starting the server right? not actually using the application. Unless there is some way to remotely start an RMI server that I am unaware of.
hmmm....I guess I need to include an error message if the client cant connect to the server if they start with no arguments. But I am just wondering is that everyone elses interpretation i.e. that when they start in server mode they are starting the RMI server only.
When you run the application with argument 'server' it means you are just starting the RMI server. I dont think there is any such thing as running the app in server mode i.e. starting the server and then running the client as well as the server should be remote and you cant start a server remotely. I know the server could be the localhost but that is why we have standalone option. That confused me initially too but I believe that the intention is that the examiner will start the rmi server on one machine and try to connect to the server from a different client machine.
My interpretation is that when starting in server mode should start up the database and the RMI registry only, and prompting for a location of the db and perhaps set rmi port number to something other than 1099. If you like you could have a small server gui for configuring the server parameters and for shutdown.
Starting with no arguments should prompt you for an ip and port number of a running server and either launch the client GUI on successful connection or throw an exception if there is no server running. [ October 12, 2006: Message edited by: Mark Smyth ]