Before discussing the details of how to start and stop the WebLogic application server, it is valuable to first talk about the different states the application server can be in while being cycled. It is possible to use either the WebLogic Administration Console or the weblogic.Admin Java tool to explicitly move the application server to a new state. The following is a listing of the various states the server can go through during its life cycle:

• shutdown

  A server is said to be in the shutdown state when it is not running. When a server is in the shutdown state, the only thing you can do is start it.

• starting

  A server is said to be in the starting state when it has been started from the command line or by a Node Manager. When in this state, the server contacts the Administration Server, retrieves its configuration information, starts all services, deploys all applications, and finally executes any startup classes.

• standby

  A server is capable of moving into the standby state if an administration port has been configured. Otherwise, the server will move directly from the starting state to the running state. When a server is in the standby state, all services are initialized and all application get deployed. With the exception of the administration port, however, all ports are closed. Therefore, teh server is said to be in hot standby and ready to go.

• running

  When a server is in the running state, it is fully operational and participating as a regular member of a WebLogic domain.

• suspending

  A server is moved into the suspending state during a graceful shutdown. This allows the server to complete any processing it may be working on to enable a clean shutdown without loss of any important state information. This state will be skipped during a forced shutdown.

• shutting down

  A server will be moved into the shutting down state after it receives a shutdown command. If the server is in the running or standby state when the shutdown command is received, it shuts down gracefully, terminating all services and executing any configured shutdown classes.

• failed

  A server is moved into the failed state when the health monitoring service determines that a critical service is not functioning properly during its life cycle. Once a server is put into the failed state, it can only be shut down. If you have enabled health monitoring, the Node Manager can automatically shut down and restart the server if it fails.

• unknown

  A server is said to be in the unknown state if it cannot be contacted.

When running the command-line examples in this article, you will need to set certain environment variables. (i.e. CLASSPATH, PATH). First, let's assume that the location of your BEA WebLogic Server installation was installed to C:\bea\weblogic81. I will refer to this location as %WL_HOME% throughout the remainder of this article.

The WebLogic Application Server includes a Command Script (or Shell Script if you are running this on UNIX) in the %WL_HOME%\server\bin directory named setWLSEnv.cmd. You can run this script to set the needed environment variables for running WebLogic Server commands.

cd C:\bea\weblogic81\server\bin
call setWLSEnv.cmd

I typically start all WebLogic servers from the command-line. When creating a domain, it is common to create a script that will start the server(s). This script will set many environment variables, start any databases (if required), and finally start the WebLogic Server. The WebLogic Server is a Java application, namely weblogic.Server, that will get called at the end of your start script.

WebLogic also includes a wrapper script that can be used to start a managed WebLogic Server. This script it typically called startManagedWebLogic.cmd and is used as follows:

Usage: startManagedWebLogic [SERVER_NAME] [ADMIN_URL]

for example:

startManagedWebLogic managedserver1 http://localhost:7001

Instead of providing the [SERVER_NAME] and [ADMIN_URL] parameters to the startManagedWebLogic.cmd, you can set the SERVER_NAME and ADMIN_URL environment variables, then run the script with no arguments.

There are two types of shutdown's that can be performed on a WebLogic server - graceful and forced. With a graceful shutdown, the server is transitioned through the following states - running, suspending, standby, shuting down, and shutdown. With a forced shutdown, the server is only transitioned through the standby and then the shutdown state.

Keep in mind that a WebLogic Server can be shut down from anywhere within it's life cycle.

One way to shutdown a WebLogic Server is using the Administration Console. Simply right-click on the server's node in the left pane and choose the "Start/Stop this server" option. From here, you can simply select the "Shutdown this server" option from the right pane.

You can also use the administration tool (a stand alone Java application) and use the SHUTDOWN option. Keep in mind that using the SHUTDOWN option will perform a graceful shutdown of the server.

In the following example, I want to shutdown a server (gracefully) named examplesServer, located on a server named bartman running on port 7001:

java weblogic.Admin
  -url t3://bartman:7001/
  -username weblogic -password weblogic
  SHUTDOWN examplesServer

You can also use the FORCESHUTDOWN option to perform a forced shutdown. Keep in mind that a forced shutdown can result in a loss of session-state information or could cause currently active transactions to rollback.

java weblogic.Admin
  -url t3://bartman:7001/
  -username weblogic -password weblogic
  FORCESHUTDOWN examplesServer

WebLogic 8.1 added several arguments to give you more control over shut down commands. The -timeout option determines the maximum number of seconds WebLogic should wait for a graceful shutdown to complete. What this does is set a limit on the completion of pending in-flight tasks. Without this argument, WebLogic will use the timeout value set for the server from the Administration Console, which by default is set to 0, which means that the server will wait indefinitely. Another argument (actually a flag) added to WebLogic 8.1 is -ignoreExistingSessions, which is used to drop all HTTP sessions immediately. Without this argument, WebLogic uses the setting defined in the Administation Console, which waits for all HTTP sesion to complete or time out. Here is an example that uses both arguments:

java weblogic.Admin
  -url t3://bartman:7001/
  -username weblogic -password weblogic
  SHUTDOWN -ignoreExistingSessions -timeout 10 examplesServer