Starting and Stopping RTserver

The rtserver command, described in detail in the TIBCO SmartSockets Utilities reference, is used to start and stop an RTserver. To start an RTserver, you can simply enter:

	On platforms that support both 32- and 64-bit, use the rtserver64 command to run the 64-bit version of RTserver.

The command must be entered with no filename extensions. Do not use rtserver.x or rtserver.exe. Using these extensions will prevent RTserver from running in the background and may cause other problems.

To start and use an RTserver, you must have a license for that RTserver. The license information must appear in the TIBCO license file or else the RTserver must be branded. For more information, see the TIBCO SmartSockets Installation Guide.

where arg_list is optional and can consist of one or more of these command arguments, separated by a space:


-check	starts the non-optimized version of RTserver, which performs additional validations and checking. If you do not specify `-check`, by default the optimized version of RTserver is started. The optimized version is faster because validation and checking is turned off. However, without the validation and checking, it is harder to debug any problems. When developing your applications or testing RTserver in your development or production environments, always start RTserver with the `-check` argument.
-command `filename`	causes RTserver to look for a file named `filename` in the current directory (the directory from which RTserver is being run) and use that as a startup command file. Values for options specified in that file override values for the same options specified in the system-level or user-level startup command file.
-help	causes RTserver to print a brief synopsis and exit.
-install -demandstart -autostart	installs the RTserver process as a Windows service. You can specify either -`demandstart` or -`autostart` as the startup mode of the service. This option is supported only on Windows systems.
-license	displays license information about your RTserver. This is useful for finding out your license number when you need to contact TIBCO Product Support. Or you can display license information by using the `rtlic` shell script, described in the TIBCO SmartSockets Utilities reference.
-no_console	specifies not to display a Windows console associated with the detached process. This argument is ignored if you specify `-no_daemon`.
-no_daemon	specifies not to start RTserver as a background process and instead run in the foreground.
-node `node_name`	starts RTserver on a remote node. A remote shell command is used with `rsh` (`remsh` on HP-UX) to start RTserver on the remote node.
`-password` `pword`	provides the password used by RTserver, along with a username, to connect to other RTservers when Basic Security is enabled. Use with the `-username` argument. `pword` size is unlimited. To specify no password, use empty quotation marks (`""`)
-server_names `names_list`	specifies the value or values to use for the option Server_Names. `names_list` is a list of logical connection names separated by commas. It allows you to override the value or values specified in a startup command file. The values you specify here must use the same syntax as any value set for the Server_Names option.
-stop	specifies that a single RTserver should be stopped. The RTclients that are connected to that RTserver continue to run, eventually detect that RTserver has stopped, and try to find or start a new RTserver.
-stop_all	specifies that one RTserver, all RTservers connected to that RTserver, and all RTclients connected to all of the above RTservers should all be stopped. The RTclients are stopped by sending them a CONTROL message with a destination subject of `_all` and containing the command `quit force`.
-stop_clients	specifies that one RTserver and all RTclients connected to that RTserver should be stopped. The RTclients are stopped by sending them a CONTROL message with a destination subject of `_all` and containing the command `quit force`.
-stop_servers	specifies that one RTserver as well as all RTservers connected to that RTserver should be stopped. The RTclients that are connected to those RTservers continue to run, eventually detect that RTserver has stopped, and try to find or start a new RTserver.
-threads n	specifies whether the RTserver should start in MP multithread mode, and the number of threads to use. If you do not specify this argument, the default is 1 and RTserver starts in normal mode. If you specify a value greater than 1, the process checks to see if this RTserver was licensed for the MP option. If yes, the RTserver is started in multi-thread mode with the number of threads you specified for `n`. If no, you receive a warning message and the RTserver is started in single-thread mode. We do not recommend setting `n` to a large number. For more information, see Server_Num_Threads. The value you specify for `-threads` overrides the value specified for the Server_Num_Threads option in any of the RTserver startup command files.
-trace_file `filename`	specifies the name of the file where the RTserver puts debug information. The default, if you do not specify this argument, is the standard output (`stdout`) which is printed to the console.
-trace_level `level`	specifies the amount of information the RTserver puts in the debug file (the file you specified in `-trace_file`). These are the values you can specify for level: `never`, no information is put in a debug file `error`, only error messages put in debug file `warning`, error and warning messages are put in debug file. This is the default setting. `info` `info_1`, provides thread information for multi-threading `info_2` `verbose` `verbose_1` `verbose_2` `debug`, provides the maximum amount of information
-uninstall	removes the RTserver as a Windows service. This option is only supported on Windows systems.
-username `name`	provides the username used by RTserver, along with a password, to connect to other RTservers when Basic Security is enabled. Use with the `-password` argument. `name` size is restricted to 64 characters.
-verbose	is the same as specifying `-trace_level info`. The `-verbose` argument is provided for backwards compatibility. The preferred method for specifying the amount of information for RTserver to provide is to use the `-trace_level` argument.
-version	causes RTserver to print version and revision levels.

Starting RTserver

RTserver runs as a background process (on OpenVMS and Windows this is known as a detached process, on MVS RTserver can run as a started task) without an interactive command interface. RTserver can be started manually from the operating system prompt, or it can be started automatically when an RTclient first tries to connect to RTserver. If you want RTclient to be able to start RTserver automatically, you must use one of the non-default start prefixes. See Start Prefix for more on start prefixes.

On Windows, if RTserver was installed as a Windows service, the proper environment variable must be set before an RTclient can automatically start an RTserver when it attempts to connect. Set the RTSERVER_CMD environment variable to:

	Automatic starting of 64-bit RTserver on Windows is not supported at this time.

For details on invoking RTserver, see the rtserver reference information in the TIBCO SmartSockets Utilities reference.

The first thing you must do to begin running RTserver is to change from the current working directory to the one that contains the RTserver command file, rtserver.cm. If no rtserver.cm file is needed, then RTserver can be started from any directory.

To start RTserver on UNIX systems, type this command at the operating system prompt:

You should see the SmartSockets banner information and a line saying that the RTserver was started successfully.

To start RTserver on Windows systems, go to the Start menu, select Programs, and select the SmartSockets program folder. Select RTserver. Or, type this command at the SmartSockets command prompt:

If RTserver started successfully, the window is automatically minimized. Check your bottom bar for the RTserver process and click on it to expand the window. In the window, you should see a message saying that the RTserver was started successfully.

Stopping RTserver

	On platforms that support both 32- and 64-bit, use the rtserver64 command to run the 64-bit version of RTserver.

To stop a running RTserver process, use the rtserver command with the -stop argument on the computer that RTserver is running on (add the command-line arguments -server_names node to stop RTserver on a remote node). This mode connects to a running RTserver and requests that RTserver gracefully shut down. It should normally never be necessary to stop and restart RTserver.

When RTserver receives a stop request, it first uses the value of the option Enable_Stop_Msgs to check if stopping RTserver is enabled, and if allowed, RTserver then exits. Note that the rtserver -stop command shuts down RTserver even if there is a project running. To stop RTserver and all RTclients that are connected to that RTserver, use rtserver with the -stop_clients argument. To stop RTserver and all RTserver processes that are connected to that RTserver, use rtserver with the -stop_servers argument. To stop RTserver, all RTclients that are connected to that RTserver, and all RTserver processes that are connected to that RTserver, use rtserver with the -stop_all argument. See Enable_Stop_Msgs for details on how to properly configure RTserver shutdown security.